Repository navigation
test(acp): align Stop fixture with the release-source mock agent - #6
Conversation
|
@coderabbitai review Please review this fixture-only release repair. It preserves every assertion and timeout; the selected environment variable must produce an active tool in both automation and upstream release snapshots. |
Action performed
|
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
Codex Review: Didn't find any major issues. Can't wait for the next one! Reviewed commit: ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
If Codex has suggestions, it will comment; otherwise it will react with 👍. Codex can also answer questions or update the PR. Try commenting "@codex address that feedback". |
|
Both focused Stop scenarios now pass locally on the automation snapshot (active-command false/true), with unchanged assertions and timeout. Hosted build/typecheck, package tests and most server shards are also green; the last shard and CodeRabbit review are still running. The release-source fixture is being checked separately before the next publication attempt. |
Thread transfer impact✅ Thread transfer remains within every enforced ceiling.
Baseline: unavailable · PR result: Scenario and decoded snapshot size10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.
Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed. |
|
No actionable comments were generated in the recent review.
|
| Layer / File(s) | Summary |
|---|---|
Test fixture and carried fix reference apps/server/src/orchestration-v2/Adapters/AcpAdapterV2.test.ts, scripts/personal-release.ts |
The test sets T3_ACP_ACTIVE_TOOL_OUTCOME to "hang" when activeBackgroundTool is true. The carried fix entry now uses commit bbf3fcd21d6832c03a616dc5a09b36274ebece0c. |
Estimated code review effort: 1 (Trivial) | ~3 minutes
Change: Other
Possibly related issues
- Issue 4 — Reports a personal update without an installed-version change. This PR updates an ACP test fixture associated with a withheld personal release publication, but the issue does not confirm that the fixture caused the hold.
Possibly related PRs
- Microck/t3code#5 — Updated ACP Stop lifecycle behavior and carried ACP patch references; this PR updates the related test fixture and carried fix entry.
Suggested reviewers: t3dotgg
Merge Risk: 
Minimal · up to 7ee75
-
The fixture and release-reference changes are ready to merge with no identified actionable risk.
Pre-merge checks: 
4
-
Full details: Description check
-
Explanation
The description clearly covers the problem, the fix, and validation status. It omits an explicit Scope and approval section and does not state why this focused fix qualifies without prior approval, but it is otherwise mostly complete.
-
Finishing Touches
-
Generate docstrings
-
- Commit to this branch
- Create a new PR
Generate unit tests (beta)
-
- Commit to this branch
- Create a new PR
-
- Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.
|
Release-source validation now completed locally: all 127 core ACP tests pass on failed release source 526263b with only this fixture variable correction. The previously timing-out active-command Stop scenario passes with its original assertions and timeout. Automation-snapshot focused scenarios also pass (2/2); hosted CI, Codex and CodeRabbit are clean on 7ee7559. No runtime code changed. |
Problem
Personal release run 41275527208 withheld publication: 126/127 core ACP tests passed, but the active-command Stop test waited for an event the newer mock agent never emitted. Upstream removed T3_ACP_EMIT_ACTIVE_TOOL_THEN_HANG; the automation snapshot still supports it.
Fix
Use the shared T3_ACP_ACTIVE_TOOL_OUTCOME=hang scenario in both snapshots. Both agents emit the same pending/in-progress command and keep the prompt alive until confirmed Stop. All assertions, timeouts and release gates remain unchanged. Final cumulative carried patch: bbf3fcd. No runtime behavior or personal patches changed.
Validation
Compared the fixture paths in failed source 526263b and automation source. Focused tests and current-head CI/reviews pending; draft until validated. Full release-source validation remains a publication gate.
AI assistance
Verity, OpenAI AI assistant with coding tools. Exact model identifier is not exposed, omitted under owner-approved exception. Automated testing only; no human testing claimed.