Addresses PR review comments:
- Move clearActiveEmbeddedRun before flushPendingToolResultsAfterIdle so
waitForEmbeddedPiRunEnd resolves within the 15s budget used by
session-reset-service.ts and commands-compact.ts (flush can take up to 30s)
- queueHandle guard (if check) prevents issues when uninitialized
- runCleanupSteps already wraps each step in try/catch, so flush failures
cannot prevent the clear from running (and vice versa)
When parallel_tool_calls is enabled and sessions_yield runs concurrently with
sessions_spawn, the embedded run was marked ended (clearActiveEmbeddedRun)
before flushPendingToolResultsAfterIdle completed. This caused the announce
flow's waitForEmbeddedPiRunEnd to resolve prematurely — before spawn
registrations committed to the subagent registry — leading to
countPendingDescendantRuns returning 0 and a premature completion announce.
Fix: move clearActiveEmbeddedRun after flushPendingToolResultsAfterIdle so the
embedded run stays active until all concurrent tool executions finish. Also fix
the silent catch block in runSubagentAnnounceFlow to defer (return false)
instead of silently proceeding when the registry check fails.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
- Added a test to ensure no warnings for legacy Brave config when bundled web search allowlist compatibility is applied.
- Updated validation logic to incorporate compatibility configuration for bundled web search plugins.
- Refactored the ensureRegistry function to utilize the new compatibility handling.
* test: align extension runtime mocks with plugin-sdk
Update stale extension tests to mock the plugin-sdk runtime barrels that production code now imports, and harden the Signal tool-result harness around system-event assertions so the channels lane matches current extension boundaries.
Regeneration-Prompt: |
Verify the failing channels-lane tests against current origin/main in an isolated worktree before changing anything. If the failures reproduce on main, keep the fix test-only unless production behavior is clearly wrong. Recent extension refactors moved Telegram, WhatsApp, and Signal code onto plugin-sdk runtime barrels, so update stale tests that still mock old core module paths to intercept the seams production code now uses. For Signal reaction notifications, avoid brittle assertions that depend on shared queued system-event state when a direct harness spy on enqueue behavior is sufficient. Preserve scope: only touch the failing tests and their local harness, then rerun the reproduced targeted tests plus the full channels lane and repo check gate.
* test: fix extension test drift on main
* fix: lazy-load bundled web search plugin registry
* test: make matrix sweeper failure injection portable
* fix: split heavy matrix runtime-api seams
* fix: simplify bundled web search id lookup
* test: tolerate windows env key casing
Reuse pi-ai's Anthropic client injection seam for streaming, and add
the OpenClaw-side provider discovery, auth, model catalog, and tests
needed to expose anthropic-vertex cleanly.
Signed-off-by: sallyom <somalley@redhat.com>
