openclaw/src/agents/subagent-registry.ts

941 lines
26 KiB
TypeScript
Raw Normal View History

import { loadConfig } from "../config/config.js";
import { callGateway } from "../gateway/call.js";
import { onAgentEvent } from "../infra/agent-events.js";
import { defaultRuntime } from "../runtime.js";
2026-01-17 05:48:34 +00:00
import { type DeliveryContext, normalizeDeliveryContext } from "../utils/delivery-context.js";
2026-02-14 20:54:31 +01:00
import { resetAnnounceQueuesForTests } from "./subagent-announce-queue.js";
import { runSubagentAnnounceFlow, type SubagentRunOutcome } from "./subagent-announce.js";
import {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
SUBAGENT_ENDED_OUTCOME_KILLED,
SUBAGENT_ENDED_REASON_COMPLETE,
SUBAGENT_ENDED_REASON_ERROR,
SUBAGENT_ENDED_REASON_KILLED,
type SubagentLifecycleEndedReason,
} from "./subagent-lifecycle-events.js";
import {
resolveCleanupCompletionReason,
resolveDeferredCleanupDecision,
} from "./subagent-registry-cleanup.js";
import {
emitSubagentEndedHookOnce,
resolveLifecycleOutcomeFromRunOutcome,
runOutcomesEqual,
} from "./subagent-registry-completion.js";
import {
countActiveDescendantRunsFromRuns,
countActiveRunsForSessionFromRuns,
findRunIdsByChildSessionKeyFromRuns,
listDescendantRunsForRequesterFromRuns,
listRunsForRequesterFromRuns,
resolveRequesterForChildSessionFromRuns,
} from "./subagent-registry-queries.js";
import {
getSubagentRunsSnapshotForRead,
persistSubagentRunsToDisk,
restoreSubagentRunsFromDisk,
} from "./subagent-registry-state.js";
import type { SubagentRunRecord } from "./subagent-registry.types.js";
import { resolveAgentTimeoutMs } from "./timeout.js";
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
export type { SubagentRunRecord } from "./subagent-registry.types.js";
const subagentRuns = new Map<string, SubagentRunRecord>();
let sweeper: NodeJS.Timeout | null = null;
let listenerStarted = false;
let listenerStop: (() => void) | null = null;
2026-01-23 00:59:44 +00:00
// Use var to avoid TDZ when init runs across circular imports during bootstrap.
var restoreAttempted = false;
const SUBAGENT_ANNOUNCE_TIMEOUT_MS = 120_000;
const MIN_ANNOUNCE_RETRY_DELAY_MS = 1_000;
const MAX_ANNOUNCE_RETRY_DELAY_MS = 8_000;
/**
* Maximum number of announce delivery attempts before giving up.
* Prevents infinite retry loops when `runSubagentAnnounceFlow` repeatedly
* returns `false` due to stale state or transient conditions (#18264).
*/
const MAX_ANNOUNCE_RETRY_COUNT = 3;
/**
* Announce entries older than this are force-expired even if delivery never
* succeeded. Guards against stale registry entries surviving gateway restarts.
*/
const ANNOUNCE_EXPIRY_MS = 5 * 60_000; // 5 minutes
function resolveAnnounceRetryDelayMs(retryCount: number) {
const boundedRetryCount = Math.max(0, Math.min(retryCount, 10));
// retryCount is "attempts already made", so retry #1 waits 1s, then 2s, 4s...
const backoffExponent = Math.max(0, boundedRetryCount - 1);
const baseDelay = MIN_ANNOUNCE_RETRY_DELAY_MS * 2 ** backoffExponent;
return Math.min(baseDelay, MAX_ANNOUNCE_RETRY_DELAY_MS);
}
function logAnnounceGiveUp(entry: SubagentRunRecord, reason: "retry-limit" | "expiry") {
const retryCount = entry.announceRetryCount ?? 0;
const endedAgoMs =
typeof entry.endedAt === "number" ? Math.max(0, Date.now() - entry.endedAt) : undefined;
const endedAgoLabel = endedAgoMs != null ? `${Math.round(endedAgoMs / 1000)}s` : "n/a";
defaultRuntime.log(
`[warn] Subagent announce give up (${reason}) run=${entry.runId} child=${entry.childSessionKey} requester=${entry.requesterSessionKey} retries=${retryCount} endedAgo=${endedAgoLabel}`,
);
}
function persistSubagentRuns() {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
persistSubagentRunsToDisk(subagentRuns);
}
const resumedRuns = new Set<string>();
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const endedHookInFlightRunIds = new Set<string>();
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
function suppressAnnounceForSteerRestart(entry?: SubagentRunRecord) {
return entry?.suppressAnnounceReason === "steer-restart";
}
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
function shouldKeepThreadBindingAfterRun(params: {
entry: SubagentRunRecord;
reason: SubagentLifecycleEndedReason;
}) {
if (params.reason === SUBAGENT_ENDED_REASON_KILLED) {
return false;
}
return params.entry.spawnMode === "session";
}
function shouldEmitEndedHookForRun(params: {
entry: SubagentRunRecord;
reason: SubagentLifecycleEndedReason;
}) {
return !shouldKeepThreadBindingAfterRun(params);
}
async function emitSubagentEndedHookForRun(params: {
entry: SubagentRunRecord;
reason?: SubagentLifecycleEndedReason;
sendFarewell?: boolean;
accountId?: string;
}) {
const reason = params.reason ?? params.entry.endedReason ?? SUBAGENT_ENDED_REASON_COMPLETE;
const outcome = resolveLifecycleOutcomeFromRunOutcome(params.entry.outcome);
const error = params.entry.outcome?.status === "error" ? params.entry.outcome.error : undefined;
await emitSubagentEndedHookOnce({
entry: params.entry,
reason,
sendFarewell: params.sendFarewell,
accountId: params.accountId ?? params.entry.requesterOrigin?.accountId,
outcome,
error,
inFlightRunIds: endedHookInFlightRunIds,
persist: persistSubagentRuns,
});
}
async function completeSubagentRun(params: {
runId: string;
endedAt?: number;
outcome: SubagentRunOutcome;
reason: SubagentLifecycleEndedReason;
sendFarewell?: boolean;
accountId?: string;
triggerCleanup: boolean;
}) {
const entry = subagentRuns.get(params.runId);
if (!entry) {
return;
}
let mutated = false;
const endedAt = typeof params.endedAt === "number" ? params.endedAt : Date.now();
if (entry.endedAt !== endedAt) {
entry.endedAt = endedAt;
mutated = true;
}
if (!runOutcomesEqual(entry.outcome, params.outcome)) {
entry.outcome = params.outcome;
mutated = true;
}
if (entry.endedReason !== params.reason) {
entry.endedReason = params.reason;
mutated = true;
}
if (mutated) {
persistSubagentRuns();
}
const suppressedForSteerRestart = suppressAnnounceForSteerRestart(entry);
const shouldEmitEndedHook =
!suppressedForSteerRestart &&
shouldEmitEndedHookForRun({
entry,
reason: params.reason,
});
const shouldDeferEndedHook =
shouldEmitEndedHook &&
params.triggerCleanup &&
entry.expectsCompletionMessage === true &&
!suppressedForSteerRestart;
if (!shouldDeferEndedHook && shouldEmitEndedHook) {
await emitSubagentEndedHookForRun({
entry,
reason: params.reason,
sendFarewell: params.sendFarewell,
accountId: params.accountId,
});
}
if (!params.triggerCleanup) {
return;
}
if (suppressedForSteerRestart) {
return;
}
startSubagentAnnounceCleanupFlow(params.runId, entry);
}
function startSubagentAnnounceCleanupFlow(runId: string, entry: SubagentRunRecord): boolean {
if (!beginSubagentCleanup(runId)) {
return false;
}
const requesterOrigin = normalizeDeliveryContext(entry.requesterOrigin);
void runSubagentAnnounceFlow({
childSessionKey: entry.childSessionKey,
childRunId: entry.runId,
requesterSessionKey: entry.requesterSessionKey,
requesterOrigin,
requesterDisplayKey: entry.requesterDisplayKey,
task: entry.task,
timeoutMs: SUBAGENT_ANNOUNCE_TIMEOUT_MS,
cleanup: entry.cleanup,
waitForCompletion: false,
startedAt: entry.startedAt,
endedAt: entry.endedAt,
label: entry.label,
outcome: entry.outcome,
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
spawnMode: entry.spawnMode,
expectsCompletionMessage: entry.expectsCompletionMessage,
}).then((didAnnounce) => {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
void finalizeSubagentCleanup(runId, entry.cleanup, didAnnounce);
});
return true;
}
function resumeSubagentRun(runId: string) {
if (!runId || resumedRuns.has(runId)) {
return;
}
const entry = subagentRuns.get(runId);
if (!entry) {
return;
}
if (entry.cleanupCompletedAt) {
return;
}
// Skip entries that have exhausted their retry budget or expired (#18264).
if ((entry.announceRetryCount ?? 0) >= MAX_ANNOUNCE_RETRY_COUNT) {
logAnnounceGiveUp(entry, "retry-limit");
entry.cleanupCompletedAt = Date.now();
persistSubagentRuns();
return;
}
if (typeof entry.endedAt === "number" && Date.now() - entry.endedAt > ANNOUNCE_EXPIRY_MS) {
logAnnounceGiveUp(entry, "expiry");
entry.cleanupCompletedAt = Date.now();
persistSubagentRuns();
return;
}
const now = Date.now();
const delayMs = resolveAnnounceRetryDelayMs(entry.announceRetryCount ?? 0);
const earliestRetryAt = (entry.lastAnnounceRetryAt ?? 0) + delayMs;
if (
entry.expectsCompletionMessage === true &&
entry.lastAnnounceRetryAt &&
now < earliestRetryAt
) {
const waitMs = Math.max(1, earliestRetryAt - now);
setTimeout(() => {
resumeSubagentRun(runId);
}, waitMs).unref?.();
resumedRuns.add(runId);
return;
}
if (typeof entry.endedAt === "number" && entry.endedAt > 0) {
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
if (suppressAnnounceForSteerRestart(entry)) {
resumedRuns.add(runId);
return;
}
if (!startSubagentAnnounceCleanupFlow(runId, entry)) {
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
return;
}
resumedRuns.add(runId);
return;
}
// Wait for completion again after restart.
const cfg = loadConfig();
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
const waitTimeoutMs = resolveSubagentWaitTimeoutMs(cfg, entry.runTimeoutSeconds);
void waitForSubagentCompletion(runId, waitTimeoutMs);
resumedRuns.add(runId);
}
function restoreSubagentRunsOnce() {
if (restoreAttempted) {
return;
}
restoreAttempted = true;
try {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const restoredCount = restoreSubagentRunsFromDisk({
runs: subagentRuns,
mergeOnly: true,
});
if (restoredCount === 0) {
return;
}
// Resume pending work.
ensureListener();
if ([...subagentRuns.values()].some((entry) => entry.archiveAtMs)) {
startSweeper();
}
for (const runId of subagentRuns.keys()) {
resumeSubagentRun(runId);
}
} catch {
// ignore restore failures
}
}
function resolveArchiveAfterMs(cfg?: ReturnType<typeof loadConfig>) {
const config = cfg ?? loadConfig();
const minutes = config.agents?.defaults?.subagents?.archiveAfterMinutes ?? 60;
if (!Number.isFinite(minutes) || minutes <= 0) {
return undefined;
}
return Math.max(1, Math.floor(minutes)) * 60_000;
}
function resolveSubagentWaitTimeoutMs(
cfg: ReturnType<typeof loadConfig>,
runTimeoutSeconds?: number,
) {
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
return resolveAgentTimeoutMs({ cfg, overrideSeconds: runTimeoutSeconds ?? 0 });
}
function startSweeper() {
if (sweeper) {
return;
}
sweeper = setInterval(() => {
void sweepSubagentRuns();
}, 60_000);
sweeper.unref?.();
}
function stopSweeper() {
if (!sweeper) {
return;
}
clearInterval(sweeper);
sweeper = null;
}
async function sweepSubagentRuns() {
const now = Date.now();
let mutated = false;
for (const [runId, entry] of subagentRuns.entries()) {
if (!entry.archiveAtMs || entry.archiveAtMs > now) {
continue;
}
subagentRuns.delete(runId);
mutated = true;
try {
await callGateway({
method: "sessions.delete",
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
params: {
key: entry.childSessionKey,
deleteTranscript: true,
emitLifecycleHooks: false,
},
timeoutMs: 10_000,
});
} catch {
// ignore
}
}
if (mutated) {
persistSubagentRuns();
}
if (subagentRuns.size === 0) {
stopSweeper();
}
}
function ensureListener() {
if (listenerStarted) {
return;
}
listenerStarted = true;
listenerStop = onAgentEvent((evt) => {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
void (async () => {
if (!evt || evt.stream !== "lifecycle") {
return;
}
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const entry = subagentRuns.get(evt.runId);
if (!entry) {
return;
}
const phase = evt.data?.phase;
if (phase === "start") {
const startedAt = typeof evt.data?.startedAt === "number" ? evt.data.startedAt : undefined;
if (startedAt) {
entry.startedAt = startedAt;
persistSubagentRuns();
}
return;
}
if (phase !== "end" && phase !== "error") {
return;
}
const endedAt = typeof evt.data?.endedAt === "number" ? evt.data.endedAt : Date.now();
const error = typeof evt.data?.error === "string" ? evt.data.error : undefined;
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const outcome: SubagentRunOutcome =
phase === "error"
? { status: "error", error }
: evt.data?.aborted
? { status: "timeout" }
: { status: "ok" };
await completeSubagentRun({
runId: evt.runId,
endedAt,
outcome,
reason: phase === "error" ? SUBAGENT_ENDED_REASON_ERROR : SUBAGENT_ENDED_REASON_COMPLETE,
sendFarewell: true,
accountId: entry.requesterOrigin?.accountId,
triggerCleanup: true,
});
})();
});
}
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
async function finalizeSubagentCleanup(
runId: string,
cleanup: "delete" | "keep",
didAnnounce: boolean,
) {
const entry = subagentRuns.get(runId);
if (!entry) {
return;
}
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
if (didAnnounce) {
const completionReason = resolveCleanupCompletionReason(entry);
await emitCompletionEndedHookIfNeeded(entry, completionReason);
completeCleanupBookkeeping({
runId,
entry,
cleanup,
completedAt: Date.now(),
});
return;
}
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const now = Date.now();
const deferredDecision = resolveDeferredCleanupDecision({
entry,
now,
activeDescendantRuns: Math.max(0, countActiveDescendantRuns(entry.childSessionKey)),
announceExpiryMs: ANNOUNCE_EXPIRY_MS,
maxAnnounceRetryCount: MAX_ANNOUNCE_RETRY_COUNT,
deferDescendantDelayMs: MIN_ANNOUNCE_RETRY_DELAY_MS,
resolveAnnounceRetryDelayMs,
});
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
if (deferredDecision.kind === "defer-descendants") {
entry.lastAnnounceRetryAt = now;
entry.cleanupHandled = false;
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
resumedRuns.delete(runId);
persistSubagentRuns();
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
setTimeout(() => {
resumeSubagentRun(runId);
}, deferredDecision.delayMs).unref?.();
return;
}
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
if (deferredDecision.retryCount != null) {
entry.announceRetryCount = deferredDecision.retryCount;
entry.lastAnnounceRetryAt = now;
}
if (deferredDecision.kind === "give-up") {
const completionReason = resolveCleanupCompletionReason(entry);
await emitCompletionEndedHookIfNeeded(entry, completionReason);
logAnnounceGiveUp(entry, deferredDecision.reason);
completeCleanupBookkeeping({
runId,
entry,
cleanup: "keep",
completedAt: now,
});
return;
}
// Allow retry on the next wake if announce was deferred or failed.
entry.cleanupHandled = false;
resumedRuns.delete(runId);
persistSubagentRuns();
if (deferredDecision.resumeDelayMs == null) {
return;
}
setTimeout(() => {
resumeSubagentRun(runId);
}, deferredDecision.resumeDelayMs).unref?.();
}
async function emitCompletionEndedHookIfNeeded(
entry: SubagentRunRecord,
reason: SubagentLifecycleEndedReason,
) {
if (
entry.expectsCompletionMessage === true &&
shouldEmitEndedHookForRun({
entry,
reason,
})
) {
await emitSubagentEndedHookForRun({
entry,
reason,
sendFarewell: true,
});
}
}
function completeCleanupBookkeeping(params: {
runId: string;
entry: SubagentRunRecord;
cleanup: "delete" | "keep";
completedAt: number;
}) {
if (params.cleanup === "delete") {
subagentRuns.delete(params.runId);
persistSubagentRuns();
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
retryDeferredCompletedAnnounces(params.runId);
return;
}
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
params.entry.cleanupCompletedAt = params.completedAt;
persistSubagentRuns();
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
retryDeferredCompletedAnnounces(params.runId);
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
}
function retryDeferredCompletedAnnounces(excludeRunId?: string) {
const now = Date.now();
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
for (const [runId, entry] of subagentRuns.entries()) {
if (excludeRunId && runId === excludeRunId) {
continue;
}
if (typeof entry.endedAt !== "number") {
continue;
}
if (entry.cleanupCompletedAt || entry.cleanupHandled) {
continue;
}
if (suppressAnnounceForSteerRestart(entry)) {
continue;
}
// Force-expire announces that have been pending too long (#18264).
const endedAgo = now - (entry.endedAt ?? now);
if (endedAgo > ANNOUNCE_EXPIRY_MS) {
logAnnounceGiveUp(entry, "expiry");
entry.cleanupCompletedAt = now;
persistSubagentRuns();
continue;
}
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
resumedRuns.delete(runId);
resumeSubagentRun(runId);
}
}
function beginSubagentCleanup(runId: string) {
const entry = subagentRuns.get(runId);
if (!entry) {
return false;
}
if (entry.cleanupCompletedAt) {
return false;
}
if (entry.cleanupHandled) {
return false;
}
entry.cleanupHandled = true;
persistSubagentRuns();
return true;
}
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
export function markSubagentRunForSteerRestart(runId: string) {
const key = runId.trim();
if (!key) {
return false;
}
const entry = subagentRuns.get(key);
if (!entry) {
return false;
}
if (entry.suppressAnnounceReason === "steer-restart") {
return true;
}
entry.suppressAnnounceReason = "steer-restart";
persistSubagentRuns();
return true;
}
export function clearSubagentRunSteerRestart(runId: string) {
const key = runId.trim();
if (!key) {
return false;
}
const entry = subagentRuns.get(key);
if (!entry) {
return false;
}
if (entry.suppressAnnounceReason !== "steer-restart") {
return true;
}
entry.suppressAnnounceReason = undefined;
persistSubagentRuns();
// If the interrupted run already finished while suppression was active, retry
// cleanup now so completion output is not lost when restart dispatch fails.
resumedRuns.delete(key);
if (typeof entry.endedAt === "number" && !entry.cleanupCompletedAt) {
resumeSubagentRun(key);
}
return true;
}
export function replaceSubagentRunAfterSteer(params: {
previousRunId: string;
nextRunId: string;
fallback?: SubagentRunRecord;
runTimeoutSeconds?: number;
}) {
const previousRunId = params.previousRunId.trim();
const nextRunId = params.nextRunId.trim();
if (!previousRunId || !nextRunId) {
return false;
}
const previous = subagentRuns.get(previousRunId);
const source = previous ?? params.fallback;
if (!source) {
return false;
}
if (previousRunId !== nextRunId) {
subagentRuns.delete(previousRunId);
resumedRuns.delete(previousRunId);
}
const now = Date.now();
const cfg = loadConfig();
const archiveAfterMs = resolveArchiveAfterMs(cfg);
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const spawnMode = source.spawnMode === "session" ? "session" : "run";
const archiveAtMs =
spawnMode === "session" ? undefined : archiveAfterMs ? now + archiveAfterMs : undefined;
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
const runTimeoutSeconds = params.runTimeoutSeconds ?? source.runTimeoutSeconds ?? 0;
const waitTimeoutMs = resolveSubagentWaitTimeoutMs(cfg, runTimeoutSeconds);
const next: SubagentRunRecord = {
...source,
runId: nextRunId,
startedAt: now,
endedAt: undefined,
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
endedReason: undefined,
endedHookEmittedAt: undefined,
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
outcome: undefined,
cleanupCompletedAt: undefined,
cleanupHandled: false,
suppressAnnounceReason: undefined,
announceRetryCount: undefined,
lastAnnounceRetryAt: undefined,
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
spawnMode,
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
archiveAtMs,
runTimeoutSeconds,
};
subagentRuns.set(nextRunId, next);
ensureListener();
persistSubagentRuns();
if (archiveAtMs) {
startSweeper();
}
void waitForSubagentCompletion(nextRunId, waitTimeoutMs);
return true;
}
export function registerSubagentRun(params: {
runId: string;
childSessionKey: string;
requesterSessionKey: string;
requesterOrigin?: DeliveryContext;
requesterDisplayKey: string;
task: string;
cleanup: "delete" | "keep";
feat(sessions): expose label in sessions.list and support label lookup in sessions_send - Add `label` field to session entries and expose it in `sessions.list` - Display label column in the web UI sessions table - Support `label` parameter in `sessions_send` for lookup by label instead of sessionKey - `sessions.patch`: Accept and store `label` field - `sessions.list`: Return `label` in session entries - `sessions_spawn`: Pass label through to registry and announce flow - `sessions_send`: Accept optional `label` param, lookup session by label if sessionKey not provided - `agent` method: Accept `label` and `spawnedBy` params (stored in session entry) - Add `label` column to sessions table in web UI - Changed session store writes to merge with existing entry (`{ ...existing, ...new }`) to preserve fields like `label` that might be set separately We attempted to implement label persistence "properly" by passing the label through the `agent` call and storing it during session initialization. However, the auto-reply flow has multiple write points that overwrite the session entry, and making all of them merge-aware proved unreliable. The working solution patches the label in the `finally` block of `runSubagentAnnounceFlow`, after all other session writes complete. This is a workaround but robust - the patch happens at the very end, just before potential cleanup. A future refactor could make session writes consistently merge-based, which would allow the cleaner approach of setting label at spawn time. ```typescript // Spawn with label sessions_spawn({ task: "...", label: "my-worker" }) // Later, find by label sessions_send({ label: "my-worker", message: "continue..." }) // Or use sessions_list to see labels sessions_list() // includes label field in response ```
2026-01-08 23:17:08 +00:00
label?: string;
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
model?: string;
runTimeoutSeconds?: number;
expectsCompletionMessage?: boolean;
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
spawnMode?: "run" | "session";
}) {
const now = Date.now();
const cfg = loadConfig();
const archiveAfterMs = resolveArchiveAfterMs(cfg);
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const spawnMode = params.spawnMode === "session" ? "session" : "run";
const archiveAtMs =
spawnMode === "session" ? undefined : archiveAfterMs ? now + archiveAfterMs : undefined;
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
const runTimeoutSeconds = params.runTimeoutSeconds ?? 0;
const waitTimeoutMs = resolveSubagentWaitTimeoutMs(cfg, runTimeoutSeconds);
const requesterOrigin = normalizeDeliveryContext(params.requesterOrigin);
subagentRuns.set(params.runId, {
runId: params.runId,
childSessionKey: params.childSessionKey,
requesterSessionKey: params.requesterSessionKey,
requesterOrigin,
requesterDisplayKey: params.requesterDisplayKey,
task: params.task,
cleanup: params.cleanup,
expectsCompletionMessage: params.expectsCompletionMessage,
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
spawnMode,
feat(sessions): expose label in sessions.list and support label lookup in sessions_send - Add `label` field to session entries and expose it in `sessions.list` - Display label column in the web UI sessions table - Support `label` parameter in `sessions_send` for lookup by label instead of sessionKey - `sessions.patch`: Accept and store `label` field - `sessions.list`: Return `label` in session entries - `sessions_spawn`: Pass label through to registry and announce flow - `sessions_send`: Accept optional `label` param, lookup session by label if sessionKey not provided - `agent` method: Accept `label` and `spawnedBy` params (stored in session entry) - Add `label` column to sessions table in web UI - Changed session store writes to merge with existing entry (`{ ...existing, ...new }`) to preserve fields like `label` that might be set separately We attempted to implement label persistence "properly" by passing the label through the `agent` call and storing it during session initialization. However, the auto-reply flow has multiple write points that overwrite the session entry, and making all of them merge-aware proved unreliable. The working solution patches the label in the `finally` block of `runSubagentAnnounceFlow`, after all other session writes complete. This is a workaround but robust - the patch happens at the very end, just before potential cleanup. A future refactor could make session writes consistently merge-based, which would allow the cleaner approach of setting label at spawn time. ```typescript // Spawn with label sessions_spawn({ task: "...", label: "my-worker" }) // Later, find by label sessions_send({ label: "my-worker", message: "continue..." }) // Or use sessions_list to see labels sessions_list() // includes label field in response ```
2026-01-08 23:17:08 +00:00
label: params.label,
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
model: params.model,
runTimeoutSeconds,
createdAt: now,
startedAt: now,
archiveAtMs,
cleanupHandled: false,
});
ensureListener();
persistSubagentRuns();
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
if (archiveAtMs) {
startSweeper();
}
// Wait for subagent completion via gateway RPC (cross-process).
// The in-process lifecycle listener is a fallback for embedded runs.
void waitForSubagentCompletion(params.runId, waitTimeoutMs);
}
async function waitForSubagentCompletion(runId: string, waitTimeoutMs: number) {
try {
const timeoutMs = Math.max(1, Math.floor(waitTimeoutMs));
2026-01-31 16:48:44 +09:00
const wait = await callGateway<{
status?: string;
startedAt?: number;
endedAt?: number;
error?: string;
}>({
method: "agent.wait",
params: {
runId,
timeoutMs,
},
timeoutMs: timeoutMs + 10_000,
});
if (wait?.status !== "ok" && wait?.status !== "error" && wait?.status !== "timeout") {
return;
}
const entry = subagentRuns.get(runId);
if (!entry) {
return;
}
let mutated = false;
if (typeof wait.startedAt === "number") {
entry.startedAt = wait.startedAt;
mutated = true;
}
if (typeof wait.endedAt === "number") {
entry.endedAt = wait.endedAt;
mutated = true;
}
if (!entry.endedAt) {
entry.endedAt = Date.now();
mutated = true;
}
const waitError = typeof wait.error === "string" ? wait.error : undefined;
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const outcome: SubagentRunOutcome =
wait.status === "error"
? { status: "error", error: waitError }
: wait.status === "timeout"
? { status: "timeout" }
: { status: "ok" };
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
if (!runOutcomesEqual(entry.outcome, outcome)) {
entry.outcome = outcome;
mutated = true;
}
if (mutated) {
persistSubagentRuns();
}
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
await completeSubagentRun({
runId,
endedAt: entry.endedAt,
outcome,
reason:
wait.status === "error" ? SUBAGENT_ENDED_REASON_ERROR : SUBAGENT_ENDED_REASON_COMPLETE,
sendFarewell: true,
accountId: entry.requesterOrigin?.accountId,
triggerCleanup: true,
});
} catch {
// ignore
}
}
export function resetSubagentRegistryForTests(opts?: { persist?: boolean }) {
subagentRuns.clear();
resumedRuns.clear();
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
endedHookInFlightRunIds.clear();
2026-02-14 20:54:31 +01:00
resetAnnounceQueuesForTests();
stopSweeper();
restoreAttempted = false;
if (listenerStop) {
listenerStop();
listenerStop = null;
}
listenerStarted = false;
if (opts?.persist !== false) {
persistSubagentRuns();
}
}
export function addSubagentRunForTests(entry: SubagentRunRecord) {
subagentRuns.set(entry.runId, entry);
}
export function releaseSubagentRun(runId: string) {
const didDelete = subagentRuns.delete(runId);
if (didDelete) {
persistSubagentRuns();
}
if (subagentRuns.size === 0) {
stopSweeper();
}
}
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
function findRunIdsByChildSessionKey(childSessionKey: string): string[] {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
return findRunIdsByChildSessionKeyFromRuns(subagentRuns, childSessionKey);
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
}
export function resolveRequesterForChildSession(childSessionKey: string): {
requesterSessionKey: string;
requesterOrigin?: DeliveryContext;
} | null {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const resolved = resolveRequesterForChildSessionFromRuns(
getSubagentRunsSnapshotForRead(subagentRuns),
childSessionKey,
);
if (!resolved) {
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
return null;
}
return {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
requesterSessionKey: resolved.requesterSessionKey,
requesterOrigin: normalizeDeliveryContext(resolved.requesterOrigin),
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
};
}
export function isSubagentSessionRunActive(childSessionKey: string): boolean {
const runIds = findRunIdsByChildSessionKey(childSessionKey);
for (const runId of runIds) {
const entry = subagentRuns.get(runId);
if (!entry) {
continue;
}
if (typeof entry.endedAt !== "number") {
return true;
}
}
return false;
}
export function markSubagentRunTerminated(params: {
runId?: string;
childSessionKey?: string;
reason?: string;
}): number {
const runIds = new Set<string>();
if (typeof params.runId === "string" && params.runId.trim()) {
runIds.add(params.runId.trim());
}
if (typeof params.childSessionKey === "string" && params.childSessionKey.trim()) {
for (const runId of findRunIdsByChildSessionKey(params.childSessionKey)) {
runIds.add(runId);
}
}
if (runIds.size === 0) {
return 0;
}
const now = Date.now();
const reason = params.reason?.trim() || "killed";
let updated = 0;
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
const entriesByChildSessionKey = new Map<string, SubagentRunRecord>();
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
for (const runId of runIds) {
const entry = subagentRuns.get(runId);
if (!entry) {
continue;
}
if (typeof entry.endedAt === "number") {
continue;
}
entry.endedAt = now;
entry.outcome = { status: "error", error: reason };
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
entry.endedReason = SUBAGENT_ENDED_REASON_KILLED;
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
entry.cleanupHandled = true;
entry.cleanupCompletedAt = now;
entry.suppressAnnounceReason = "killed";
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
if (!entriesByChildSessionKey.has(entry.childSessionKey)) {
entriesByChildSessionKey.set(entry.childSessionKey, entry);
}
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
updated += 1;
}
if (updated > 0) {
persistSubagentRuns();
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
for (const entry of entriesByChildSessionKey.values()) {
void emitSubagentEndedHookOnce({
entry,
reason: SUBAGENT_ENDED_REASON_KILLED,
sendFarewell: true,
outcome: SUBAGENT_ENDED_OUTCOME_KILLED,
error: reason,
inFlightRunIds: endedHookInFlightRunIds,
persist: persistSubagentRuns,
}).catch(() => {
// Hook failures should not break termination flow.
});
}
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
}
return updated;
}
2026-01-16 21:37:11 +00:00
export function listSubagentRunsForRequester(requesterSessionKey: string): SubagentRunRecord[] {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
return listRunsForRequesterFromRuns(subagentRuns, requesterSessionKey);
2026-01-16 21:37:11 +00:00
}
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
export function countActiveRunsForSession(requesterSessionKey: string): number {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
return countActiveRunsForSessionFromRuns(
getSubagentRunsSnapshotForRead(subagentRuns),
requesterSessionKey,
);
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
}
export function countActiveDescendantRuns(rootSessionKey: string): number {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
return countActiveDescendantRunsFromRuns(
getSubagentRunsSnapshotForRead(subagentRuns),
rootSessionKey,
);
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
}
export function listDescendantRunsForRequester(rootSessionKey: string): SubagentRunRecord[] {
feat: thread-bound subagents on Discord (#21805) * docs: thread-bound subagents plan * docs: add exact thread-bound subagent implementation touchpoints * Docs: prioritize auto thread-bound subagent flow * Docs: add ACP harness thread-binding extensions * Discord: add thread-bound session routing and auto-bind spawn flow * Subagents: add focus commands and ACP/session binding lifecycle hooks * Tests: cover thread bindings, focus commands, and ACP unbind hooks * Docs: add plugin-hook appendix for thread-bound subagents * Plugins: add subagent lifecycle hook events * Core: emit subagent lifecycle hooks and decouple Discord bindings * Discord: handle subagent bind lifecycle via plugin hooks * Subagents: unify completion finalizer and split registry modules * Add subagent lifecycle events module * Hooks: fix subagent ended context key * Discord: share thread bindings across ESM and Jiti * Subagents: add persistent sessions_spawn mode for thread-bound sessions * Subagents: clarify thread intro and persistent completion copy * test(subagents): stabilize sessions_spawn lifecycle cleanup assertions * Discord: add thread-bound session TTL with auto-unfocus * Subagents: fail session spawns when thread bind fails * Subagents: cover thread session failure cleanup paths * Session: add thread binding TTL config and /session ttl controls * Tests: align discord reaction expectations * Agent: persist sessionFile for keyed subagent sessions * Discord: normalize imports after conflict resolution * Sessions: centralize sessionFile resolve/persist helper * Discord: harden thread-bound subagent session routing * Rebase: resolve upstream/main conflicts * Subagents: move thread binding into hooks and split bindings modules * Docs: add channel-agnostic subagent routing hook plan * Agents: decouple subagent routing from Discord * Discord: refactor thread-bound subagent flows * Subagents: prevent duplicate end hooks and orphaned failed sessions * Refactor: split subagent command and provider phases * Subagents: honor hook delivery target overrides * Discord: add thread binding kill switches and refresh plan doc * Discord: fix thread bind channel resolution * Routing: centralize account id normalization * Discord: clean up thread bindings on startup failures * Discord: add startup cleanup regression tests * Docs: add long-term thread-bound subagent architecture * Docs: split session binding plan and dedupe thread-bound doc * Subagents: add channel-agnostic session binding routing * Subagents: stabilize announce completion routing tests * Subagents: cover multi-bound completion routing * Subagents: suppress lifecycle hooks on failed thread bind * tests: fix discord provider mock typing regressions * docs/protocol: sync slash command aliases and delete param models * fix: add changelog entry for Discord thread-bound subagents (#21805) (thanks @onutc) --------- Co-authored-by: Shadow <hi@shadowing.dev>
2026-02-21 16:14:55 +01:00
return listDescendantRunsForRequesterFromRuns(
getSubagentRunsSnapshotForRead(subagentRuns),
rootSessionKey,
);
Agents: add nested subagent orchestration controls and reduce subagent token waste (#14447) * Agents: add subagent orchestration controls * Agents: add subagent orchestration controls (WIP uncommitted changes) * feat(subagents): add depth-based spawn gating for sub-sub-agents * feat(subagents): tool policy, registry, and announce chain for nested agents * feat(subagents): system prompt, docs, changelog for nested sub-agents * fix(subagents): prevent model fallback override, show model during active runs, and block context overflow fallback Bug 1: When a session has an explicit model override (e.g., gpt/openai-codex), the fallback candidate logic in resolveFallbackCandidates silently appended the global primary model (opus) as a backstop. On reinjection/steer with a transient error, the session could fall back to opus which has a smaller context window and crash. Fix: when storedModelOverride is set, pass fallbacksOverride ?? [] instead of undefined, preventing the implicit primary backstop. Bug 2: Active subagents showed 'model n/a' in /subagents list because resolveModelDisplay only read entry.model/modelProvider (populated after run completes). Fix: fall back to modelOverride/providerOverride fields which are populated at spawn time via sessions.patch. Bug 3: Context overflow errors (prompt too long, context_length_exceeded) could theoretically escape runEmbeddedPiAgent and be treated as failover candidates in runWithModelFallback, causing a switch to a model with a smaller context window. Fix: in runWithModelFallback, detect context overflow errors via isLikelyContextOverflowError and rethrow them immediately instead of trying the next model candidate. * fix(subagents): track spawn depth in session store and fix announce routing for nested agents * Fix compaction status tracking and dedupe overflow compaction triggers * fix(subagents): enforce depth block via session store and implement cascade kill * fix: inject group chat context into system prompt * fix(subagents): always write model to session store at spawn time * Preserve spawnDepth when agent handler rewrites session entry * fix(subagents): suppress announce on steer-restart * fix(subagents): fallback spawned session model to runtime default * fix(subagents): enforce spawn depth when caller key resolves by sessionId * feat(subagents): implement active-first ordering for numeric targets and enhance task display - Added a test to verify that subagents with numeric targets follow an active-first list ordering. - Updated `resolveSubagentTarget` to sort subagent runs based on active status and recent activity. - Enhanced task display in command responses to prevent truncation of long task descriptions. - Introduced new utility functions for compacting task text and managing subagent run states. * fix(subagents): show model for active runs via run record fallback When the spawned model matches the agent's default model, the session store's override fields are intentionally cleared (isDefault: true). The model/modelProvider fields are only populated after the run completes. This left active subagents showing 'model n/a'. Fix: store the resolved model on SubagentRunRecord at registration time, and use it as a fallback in both display paths (subagents tool and /subagents command) when the session store entry has no model info. Changes: - SubagentRunRecord: add optional model field - registerSubagentRun: accept and persist model param - sessions-spawn-tool: pass resolvedModel to registerSubagentRun - subagents-tool: pass run record model as fallback to resolveModelDisplay - commands-subagents: pass run record model as fallback to resolveModelDisplay * feat(chat): implement session key resolution and reset on sidebar navigation - Added functions to resolve the main session key and reset chat state when switching sessions from the sidebar. - Updated the `renderTab` function to handle session key changes when navigating to the chat tab. - Introduced a test to verify that the session resets to "main" when opening chat from the sidebar navigation. * fix: subagent timeout=0 passthrough and fallback prompt duplication Bug 1: runTimeoutSeconds=0 now means 'no timeout' instead of applying 600s default - sessions-spawn-tool: default to undefined (not 0) when neither timeout param is provided; use != null check so explicit 0 passes through to gateway - agent.ts: accept 0 as valid timeout (resolveAgentTimeoutMs already handles 0 → MAX_SAFE_TIMEOUT_MS) Bug 2: model fallback no longer re-injects the original prompt as a duplicate - agent.ts: track fallback attempt index; on retries use a short continuation message instead of the full original prompt since the session file already contains it from the first attempt - Also skip re-sending images on fallback retries (already in session) * feat(subagents): truncate long task descriptions in subagents command output - Introduced a new utility function to format task previews, limiting their length to improve readability. - Updated the command handler to use the new formatting function, ensuring task descriptions are truncated appropriately. - Adjusted related tests to verify that long task descriptions are now truncated in the output. * refactor(subagents): update subagent registry path resolution and improve command output formatting - Replaced direct import of STATE_DIR with a utility function to resolve the state directory dynamically. - Enhanced the formatting of command output for active and recent subagents, adding separators for better readability. - Updated related tests to reflect changes in command output structure. * fix(subagent): default sessions_spawn to no timeout when runTimeoutSeconds omitted The previous fix (75a791106) correctly handled the case where runTimeoutSeconds was explicitly set to 0 ("no timeout"). However, when models omit the parameter entirely (which is common since the schema marks it as optional), runTimeoutSeconds resolved to undefined. undefined flowed through the chain as: sessions_spawn → timeout: undefined (since undefined != null is false) → gateway agent handler → agentCommand opts.timeout: undefined → resolveAgentTimeoutMs({ overrideSeconds: undefined }) → DEFAULT_AGENT_TIMEOUT_SECONDS (600s = 10 minutes) This caused subagents to be killed at exactly 10 minutes even though the user's intent (via TOOLS.md) was for subagents to run without a timeout. Fix: default runTimeoutSeconds to 0 (no timeout) when neither runTimeoutSeconds nor timeoutSeconds is provided by the caller. Subagent spawns are long-running by design and should not inherit the 600s agent-command default timeout. * fix(subagent): accept timeout=0 in agent-via-gateway path (second 600s default) * fix: thread timeout override through getReplyFromConfig dispatch path getReplyFromConfig called resolveAgentTimeoutMs({ cfg }) with no override, always falling back to the config default (600s). Add timeoutOverrideSeconds to GetReplyOptions and pass it through as overrideSeconds so callers of the dispatch chain can specify a custom timeout (0 = no timeout). This complements the existing timeout threading in agentCommand and the cron isolated-agent runner, which already pass overrideSeconds correctly. * feat(model-fallback): normalize OpenAI Codex model references and enhance fallback handling - Added normalization for OpenAI Codex model references, specifically converting "gpt-5.3-codex" to "openai-codex" before execution. - Updated the `resolveFallbackCandidates` function to utilize the new normalization logic. - Enhanced tests to verify the correct behavior of model normalization and fallback mechanisms. - Introduced a new test case to ensure that the normalization process works as expected for various input formats. * feat(tests): add unit tests for steer failure behavior in openclaw-tools - Introduced a new test file to validate the behavior of subagents when steer replacement dispatch fails. - Implemented tests to ensure that the announce behavior is restored correctly and that the suppression reason is cleared as expected. - Enhanced the subagent registry with a new function to clear steer restart suppression. - Updated related components to support the new test scenarios. * fix(subagents): replace stop command with kill in slash commands and documentation - Updated the `/subagents` command to replace `stop` with `kill` for consistency in controlling sub-agent runs. - Modified related documentation to reflect the change in command usage. - Removed legacy timeoutSeconds references from the sessions-spawn-tool schema and tests to streamline timeout handling. - Enhanced tests to ensure correct behavior of the updated commands and their interactions. * feat(tests): add unit tests for readLatestAssistantReply function - Introduced a new test file for the `readLatestAssistantReply` function to validate its behavior with various message scenarios. - Implemented tests to ensure the function correctly retrieves the latest assistant message and handles cases where the latest message has no text. - Mocked the gateway call to simulate different message histories for comprehensive testing. * feat(tests): enhance subagent kill-all cascade tests and announce formatting - Added a new test to verify that the `kill-all` command cascades through ended parents to active descendants in subagents. - Updated the subagent announce formatting tests to reflect changes in message structure, including the replacement of "Findings:" with "Result:" and the addition of new expectations for message content. - Improved the handling of long findings and stats in the announce formatting logic to ensure concise output. - Refactored related functions to enhance clarity and maintainability in the subagent registry and tools. * refactor(subagent): update announce formatting and remove unused constants - Modified the subagent announce formatting to replace "Findings:" with "Result:" and adjusted related expectations in tests. - Removed constants for maximum announce findings characters and summary words, simplifying the announcement logic. - Updated the handling of findings to retain full content instead of truncating, ensuring more informative outputs. - Cleaned up unused imports in the commands-subagents file to enhance code clarity. * feat(tests): enhance billing error handling in user-facing text - Added tests to ensure that normal text mentioning billing plans is not rewritten, preserving user context. - Updated the `isBillingErrorMessage` and `sanitizeUserFacingText` functions to improve handling of billing-related messages. - Introduced new test cases for various scenarios involving billing messages to ensure accurate processing and output. - Enhanced the subagent announce flow to correctly manage active descendant runs, preventing premature announcements. * feat(subagent): enhance workflow guidance and auto-announcement clarity - Added a new guideline in the subagent system prompt to emphasize trust in push-based completion, discouraging busy polling for status updates. - Updated documentation to clarify that sub-agents will automatically announce their results, improving user understanding of the workflow. - Enhanced tests to verify the new guidance on avoiding polling loops and to ensure the accuracy of the updated prompts. * fix(cron): avoid announcing interim subagent spawn acks * chore: clean post-rebase imports * fix(cron): fall back to child replies when parent stays interim * fix(subagents): make active-run guidance advisory * fix(subagents): update announce flow to handle active descendants and enhance test coverage - Modified the announce flow to defer announcements when active descendant runs are present, ensuring accurate status reporting. - Updated tests to verify the new behavior, including scenarios where no fallback requester is available and ensuring proper handling of finished subagents. - Enhanced the announce formatting to include an `expectFinal` flag for better clarity in the announcement process. * fix(subagents): enhance announce flow and formatting for user updates - Updated the announce flow to provide clearer instructions for user updates based on active subagent runs and requester context. - Refactored the announcement logic to improve clarity and ensure internal context remains private. - Enhanced tests to verify the new message expectations and formatting, including updated prompts for user-facing updates. - Introduced a new function to build reply instructions based on session context, improving the overall announcement process. * fix: resolve prep blockers and changelog placement (#14447) (thanks @tyler6204) * fix: restore cron delivery-plan import after rebase (#14447) (thanks @tyler6204) * fix: resolve test failures from rebase conflicts (#14447) (thanks @tyler6204) * fix: apply formatting after rebase (#14447) (thanks @tyler6204)
2026-02-14 22:03:45 -08:00
}
export function initSubagentRegistry() {
restoreSubagentRunsOnce();
}