OpenClaw's default agent timeout is 600s (10 min), which consistently kills
long-running agent responses — especially multi-file code generation and
complex multi-tool-call workflows. This triggers:
- 'terminated' errors mid-stream (webchat shows terminated banner)
- Retry storms: same runId retries ~60s apart, then new runIds also fail
- Follow-up messages silently dropped (shouldSkipMessageByAbortCutoff)
Set the default to 86400s (24 hours) during bootstrap so all DenchClaw
installations get a sane timeout out of the box. Note: the runtime code
supports timeoutSeconds=0 (no timeout) but the config schema validation
rejects 0 (exclusiveMinimum: 0), so 86400 is the practical maximum.
Also rename ensureSubagentDefaults -> ensureAgentDefaults since the function
now configures both agent-level and subagent-level settings.
Refs:
- https://github.com/openclaw/openclaw/issues/30487
- https://github.com/openclaw/openclaw/issues/46049
- https://github.com/openclaw/openclaw/issues/30083
Add optional name, email, avatar, and denchOrgId fields to
telemetry.json. When present, all telemetry layers (CLI, web server,
web client, OpenClaw plugin) call PostHog identify() with $name,
$email, $avatar, and dench_org_id person properties.
Remove $process_person_profile:false from all layers so every install
gets a PostHog person profile. Enable session replay with masking
controlled by privacy mode (all text/inputs masked when on, nothing
masked when off).
Move identity injection from static IDENTITY.md generation at workspace seed
time to a runtime `dench-identity` OpenClaw plugin that prepends the system
prompt via the `before_prompt_build` hook. This keeps identity always current
without re-seeding, encourages dynamic skill discovery, and positions
DenchClaw as a CEO orchestrator that delegates to specialist subagents.
- Back up existing app dir before replacing (enables rollback on crash)
- Rollback to previous working runtime when new version fails readiness probe
- Early exit in waitForWebRuntime() when spawned process dies (saves 15s)
- Include web-app.err.log tail in probe failure reason for diagnostics
- Clean up backup after successful probe
Made-with: Cursor
Chat-slot agents were being persisted as durable entries in openclaw.json,
causing spurious workspace directories (e.g. chat-slot-main-1) to appear.
Only explicit workspace creation via init now creates durable agent entries.
Workspace discovery and session routing ignore chat-slot internals.
Support DENCHCLAW_DAEMONLESS=1 env var and --skip-daemon-install flag
across all commands (bootstrap, update, start, restart, stop) to skip
gateway daemon management and LaunchAgent registration in environments
without systemd/launchd.
Pre-onboard config (gateway.mode, gateway.port, plugin trust) used to call
`openclaw config set` before `openclaw onboard` created the profile, causing
"Failed to set ..." errors on fresh installs (#101). All pre-onboard config is
now staged via raw JSON writes. Hidden gateway restarts inside syncBundledPlugins
are removed; one explicit restart happens after all post-onboard config, followed
by retried health probes so bootstrap no longer falsely reports "gateway closed"
while the gateway is simply finishing startup.
extractUsageFromMessages was summing ALL assistant messages, so each
$ai_generation event reported a growing cumulative total — PostHog's
sum() then massively over-counted costs (e.g. $634 reported vs $73 actual).
Integrate version tracking for DenchClaw and OpenClaw into the telemetry system. The versions are now read from the package.json and environment variables, and are included in the PostHog client initialization and telemetry events. This enhancement allows for better monitoring and analytics of the versions in use.
On fresh Linux installs (e.g. running as root), `openclaw config set
agents.defaults.workspace` fails because the target directory doesn't
exist yet — some OpenClaw builds validate the path on disk before
accepting the value. Create the directory eagerly with mkdirSync before
the config set call.
Also surface the exit code in runOpenClawOrThrow errors when stderr is
empty, so silent failures are easier to diagnose.
Fixes#101
Set gateway mode and port before onboarding so the first daemon start succeeds, then reapply them after onboarding so wizard defaults cannot drift DenchClaw off its expected local gateway.
During bootstrap, `onboard --install-daemon` starts the gateway daemon
immediately. The gateway's startup guard requires `gateway.mode=local`
but this was only set *after* onboard completed, causing the daemon to
block with "Gateway start blocked: set gateway.mode=local" and enter a
crash loop. The web UI then fails with "Gateway WebSocket connection
failed".
Move `ensureGatewayModeLocal()` and `ensureGatewayPort()` to run before
the onboard command so the config is in place when the daemon first
starts.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
When denchclaw is launched via `npx`, npm injects environment variables
(most critically `npm_config_prefix`) that redirect `npm install -g` and
`npm ls -g` to a temporary npx-managed prefix instead of the user's real
global npm directory. This causes openclaw to be installed to a location
that doesn't persist on PATH, so subsequent commands like `denchclaw
restart` fail with "openclaw CLI not found".
Strip `npm_config_*`, `npm_package_*`, and npm lifecycle variables from
the environment passed to npm global commands (`npm ls -g`, `npm install
-g`, `npm prefix -g`) so they use the user's actual npm configuration.
- Convert sync filesystem and DuckDB operations to async across API routes,
workspace lib, and active-runs to prevent event loop blocking during tree
discovery, object lookups, and database queries
- Add "tags" field type for free-form string arrays with parse-tags utility,
TagsBadges/TagsInput UI components, filter operators, and CRM skill docs
- Preserve rich text formatting (bold, italic, code, @mentions) in user chat
messages by sending HTML alongside plain text through the transport layer
- Detect empty-stream errors, improve agent error emission, and add file
mutation queues for concurrent write safety in active-runs
- Add pre-publish standalone node_modules verification in deploy script
checking serverExternalPackages are present
- Extract syncManagedSkills and discoverWorkspaceDirs for multi-workspace
skill syncing, add ensureSeedAssets for runtime app dir
- Bump version 2.1.1 → 2.1.4
Update package versions for denchclaw and dench to 2.0.16. Refactor feedback API to convert chat lines into PostHog-compatible messages, preserving chronological order and including tool calls and results. Improve test cases to reflect changes in message handling.
Use the session key directly as the PostHog trace ID so feedback and generation events share the same trace. Extract non-assistant messages from conversations for $ai_input.
A random UUID is generated on first run and stored in ~/.openclaw-dench/telemetry.json, replacing the SHA-256 hostname hash that could theoretically be reversed.
Introduce a `noOpen` option across various web runtime commands to control automatic browser opening. Implement a prompt to ask users if they want to open the web UI, enhancing user experience and flexibility during command execution.
Enhance the installBundledPlugins function to set the plugins.load.paths configuration for posthog-analytics using OpenClaw. This addition allows for dynamic loading of plugin paths, improving plugin management during the bootstrap process.
Add extractUsageFromMessages and normalizeOutputForPostHog to pull actual
token counts and cost from OpenClaw per-message usage metadata, and convert
Anthropic-style tool call blocks to OpenAI format for PostHog AI observability.
Integrate gateway daemon restart (stop → install → start) into both the
update and start web runtime commands so the gateway stays in sync with
runtime upgrades.
Switch TraceContextManager from runId to sessionId-based indexing so
tool spans share the same trace_id as their parent generation. Extract
tool names from agent_end messages for the PostHog Tools tab. Guard
against sending $0 cost and null token counts.
Integrate PostHog LLM Analytics via a bundled OpenClaw plugin that captures
$ai_generation, $ai_span, and $ai_trace events with configurable privacy
mode (content redaction on by default). Add like/dislike feedback buttons
to the web chat UI backed by a /api/feedback route. Extend the CLI with
`telemetry privacy on|off` subcommands and fix command delegation so
telemetry subcommands aren't forwarded to OpenClaw. Harden the web runtime
installer to auto-flatten pnpm standalone deps and dereference dangling
symlinks, preventing "Cannot find module 'next'" crashes in dev. Move
plugin installation before onboard in bootstrap so the gateway starts
with plugins.allow already configured.
- New src/config/paths.test.ts: 11 tests covering resolveGatewayPort
profile-aware precedence (env > config > profile > global default)
- Extend bootstrap-external.test.ts: 15 new tests for
readExistingGatewayPort (config reading, fallback, edge cases) and
isPersistedPortAcceptable (18789 rejection guard, end-to-end composition)
- Extend web-runtime-command.test.ts: 2 new tests verifying the fallback
returns 19001 when manifest has no lastGatewayPort or is null
- Update test fixtures across all test files to expect port 19001
All 5 critical mutations verified: removing profile check, removing
18789 guard, reverting fallback, changing constant, breaking config
reader -- each caught by at least 2 tests.
The bootstrap port selection logic tried OpenClaw's default port (18789)
before the DenchClaw range (19001+). If OpenClaw was temporarily down
during bootstrap, DenchClaw would claim 18789 and persist it to config
and the LaunchAgent plist, killing OpenClaw on every subsequent restart.
- Remove the DEFAULT_GATEWAY_PORT (18789) branch from bootstrap; always
start from DENCHCLAW_GATEWAY_PORT_START (19001)
- Read previously persisted config port and reuse it (unless it is 18789)
- Extract isPersistedPortAcceptable guard to reject corrupted 18789 state
- Add DENCHCLAW_DEFAULT_GATEWAY_PORT constant and make resolveGatewayPort
profile-aware so the "dench" profile falls back to 19001
- Fix hardcoded 18789 fallback in web-runtime-command.ts
pnpm's standalone output uses symlinks that npm pack silently drops, breaking require('next') on user machines. This flattens the virtual store into a standard node_modules layout.
Always pass --skip-ui and --accept-risk to openclaw onboard so the
wizard never prompts for TUI/Web UI selection — bootstrap manages the
web UI lifecycle itself. Add a post-onboard spinner to eliminate the
silent gap while config-set calls, gateway probing, and web runtime
startup run. Fix remediation messages to use `npx denchclaw`.
- Fix bootstrap-command test: mock ensureManagedWebRuntime to probe
directly instead of requiring standalone build on disk
- Add PostHog telemetry to CLI and web app with opt-out support
- Add dench alias package (npm rejects name; kept for future use)
- Bump version to 2.0.4 and publish to npm
Refactor bootstrap to use a managed web runtime lifecycle instead of
ad-hoc standalone server spawning. The managed runtime copies packaged
Next.js assets into ~/.openclaw-dench/web-runtime/, tracks deployment
state via manifest/process metadata, and cleanly separates Dench-owned
processes from foreign listeners on the target port.
- Fix false-negative web readiness when /api/profiles returns null
activeProfile (first-run regression).
- Add `dench start` (start without updating assets), `dench stop`
(terminate only Dench-managed web server), and `dench update`
(refresh web runtime with major-version OpenClaw update gate).
- Major-version transitions (e.g. v2->v3) require mandatory OpenClaw
update; non-interactive mode fails closed without --yes.
- All lifecycle commands show the ASCII banner/logo animation.
- Deploy smoke checks now verify update/stop/start --help paths.
This commit introduces the ensureSubagentDefaults function, which sets various default configurations for subagents, including max concurrent agents, max spawn depth, and run timeout settings. The function is called during the bootstrap process to ensure these defaults are applied for the specified profile.
This commit deletes the debug.log file, which contained subagent session logs, and updates the Ironclaw identity in workspace-seed.ts to reflect the new author, changing from Kumar Abhirup to Dench.
