Repository navigation
LCORE-5254: Restrict Fullsend retro to explicit /fs-retro comments - #3074
Conversation
Prevent the retro agent from running automatically on every closed pull request. Instead, only trigger retro on explicit /fs-retro comments from humans. This avoids unnecessary retro runs and potential side effects from automated workflows. Before this change: - Retro runs on every PR close (merged or not) - Can add labels or create issues as side effects - Adds noise to closed PRs After this change: - Retro only runs on explicit /fs-retro comment (opt-in) - Manual retrospective requests still work via /fs-retro Implemented via documented CEL trigger from Fullsend config reference. https://fullsend.sh/docs/guides/user/customizing-agents#choosing-when-retro-runs Signed-off-by: Anik Bhattacharjee <anbhatta@redhat.com>
|
No actionable comments were generated in the recent review.
|
| Layer / File(s) | Summary |
|---|---|
Configure retro role trigger .fullsend/config.yaml |
The retro role triggers on comment-added transitions when the actor is human and the comment command is /fs-retro. |
Priority:
Estimated code review effort: 1 (Trivial) | ~3 minutes
Change: Bug fix
Possibly related PRs
- lightspeed-core/lightspeed-stack#3070: Changes the Fullsend
reviewrole to use a human/fs-reviewcomment trigger. - lightspeed-core/lightspeed-stack#3055: Introduces the Fullsend configuration and enables the
retrorole.
Suggested reviewers: tisnik, jrobertboos
Merge Risk: 
Minimal · up to 0cd76
-
The opt-in retro trigger is correctly configured and introduces no identified merge-blocking risk.
Pre-merge checks: 
7
-
Full details: Title check
-
Explanation
The title names the main change: the Fullsend retro now runs only on explicit /fs-retro comments. The raw summary confirms a retro role trigger that matches human-authored comments with the /fs-retro command. The objectives state the retro no longer runs automatically when a pull request closes. The title is specific, concise, and uses the ticket prefix common in this repository. It is 66 characters, which is within a normal title length.
Full details: Docstring Coverage
-
Explanation
No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentation-only edits and deletions within surviving functions are included; unchanged diff context and fully deleted functions are excluded. Analyzed 0 functions across 0 files. (1 skipped: 1 unsupported.)
Full details: Performance And Algorithmic Complexity
-
Explanation
PASSED. The pull request changes only
.fullsend/config.yamland adds a bounded CEL trigger expression. It introduces no loops, API calls, list operations, parsing work, caches, watchers, buffers, or unbounded data growth. No meaningful performance regression is present.
Full details: Security And Secret Handling
-
Explanation
PASSED. The pull request changes only
.fullsend/config.yamlby adding a human-only/fs-retrocomment trigger at lines 13-18. It adds no plaintext secret, token, endpoint, API response, log statement, command construction, path handling, Kubernetes Secret, or RH Secret. Existing workflow secret references and the WIF provider are unchanged.
-
Finishing Touches
-
🧪 Generate unit tests (beta)
-
- Create a new PR
✨ Simplify code
-
- 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.
Comment @coderabbitai help to get the list of available commands.
|
🤖 Finished Retro · ✅ Success · Started 8:06 PM UTC · Completed 8:15 PM UTC Commit: Runtime: pi · Model: opus → claude-opus-4-6 · Effort: high · Cost: $2.32 |
|
Retro for PR #3074 — Clean workflow, no proposals. PR #3074 is a human-authored, config-only change (+6 lines to
Bootstrap note: This retro itself was dispatched automatically on merge — the old config (without the retro trigger restriction) was in effect when the Broader context: The repo installed fullsend on Oct 28 (PR #3055). Within 24 hours, automatic retro runs filed 5 issues (#3063–3069). The team is calibrating agent autonomy by restricting retro (this PR, merged) and review (PR #3070, open) to manual invocation. This is a natural onboarding calibration step. Related platform concern: fullsend-ai/fullsend#8912 tracks tightening retro authorization for closed PRs. Autonomy readiness: Inconclusive — no agent review ran, and the human approved without detailed inline comments, so no agent-vs-human delta can be computed. |
Description
Prevent the retro agent from running automatically on every closed pull request. Instead, only trigger retro on explicit
/fs-retrocomments from humans. This avoids unnecessary retro runs and potential side effects from automated workflows.Before this change:
After this change:
/fs-retrocomment (opt-in)/fs-retroImplemented via documented CEL trigger from Fullsend config reference. https://fullsend.sh/docs/guides/user/customizing-agents#choosing-when-retro-runs
Type of change
pyproject.toml+uv.lock]requirements.*.txtfor Konflux]Tools used to create PR
Identify any AI code assistants used in this PR (for transparency and review context)
Related Tickets & Documents
Checklist before requesting a review
Testing
Summary by CodeRabbit
/fs-retrocommand that starts a retrospective when included in a human-authored comment during a transition.