feat(commands): wire @gittensory review / re-review command (#2163) - #4050
Conversation
…d#2163) Dispatch authorized maintainer PR comments to the existing reReviewStoredPullRequest path with classifier skips, denial audits, and completion telemetry — without mutating gate disposition. Co-authored-by: Cursor <cursoragent@cursor.com>
|
Warning 🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨🟨 ⏸️ Gittensory review result - manual review recommendedReview updated: 2026-07-07 14:08:38 UTC
⏸️ Suggested Action - Manual Review Review summary Nits — 5 non-blocking
Review context
Contributor next steps
Signal definitions
🟩 Safe / merged · 🟦 Advisory · 🟨 Held for review · 🟥 Blocked / closed 💰 Earn for open-source contributions like this. Gittensor lets GitHub contributors earn for the work they already do — register to start earning →. Checked by Gittensory, a quiet PR intelligence layer for OSS maintainers.
|
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #4050 +/- ##
=======================================
Coverage 93.56% 93.57%
=======================================
Files 378 378
Lines 35448 35485 +37
Branches 13000 13014 +14
=======================================
+ Hits 33167 33204 +37
Misses 1618 1618
Partials 663 663
🚀 New features to boost your workflow:
|
|
🚨 Contributor flagged. Click here for more info: Superagent Dashboard |
|
Superagent didn't find any vulnerabilities or security issues in this PR. |
…typecheck (#4259) Two independently-merged PRs (#4050 and #4175) both implemented the @gittensory review command, leaving two colliding declarations of maybeProcessReviewCommand/recordReviewCommandSkip and two dispatch call sites -- a hard tsc "Duplicate function implementation" error that blocks every PR's typecheck/validate check on main. Keeps the earlier, more complete implementation (which also wires the paired resume command and a real hasAutoreviewPausedMarker fix), and folds in two things only the later, now-removed duplicate got right: - needsMinerDetection: true on the authorization call. "review" is deliberately widened to confirmed_miner (self-rerun precedent, same as review-now), so without this flag a confirmed miner re-triggering review on their own PR had no other role to match and was always denied. - A resolveAgentActionMode pause/dry-run gate before dispatching, matching every other action command (pause/resolve/explain/ gate-override/generate-tests). The kept implementation had no such gate at all, so review always dispatched live regardless of a maintainer's global pause or dry-run setting. Removes the redundant duplicate test suite for the same command and adds regression coverage for both fixes above.
Summary
maybeProcessReviewCommandfor@gittensory review/re-review: sharedclassifyPrCommandRequestguards,authorizePrActionActorfor thereviewcommand (including confirmed-miner detection), then dispatch to the existingreReviewStoredPullRequestpath withforce: true— no gate-disposition mutation (PR command surface — @gittensory review / pause / resume / resolve / help / configuration / explain #1960).github_app.review_command_{skipped,denied,completed}audit events and matching product-usage telemetry, mirroring the gate-override skip/denied/completed shape.Test plan
@gittensory reviewdispatches re-review and recordsreview_command_completedwithout changinggate_check_modeor emittinggate_overridden.@gittensory re-reviewalias resolves to the same handler.review_command_denied.review_command_skippedwith classifier reasons.cached_pr_missingskip.agentPaused/agentDryRunskip without dispatching re-review.headSha,commentId,repoFullName) covered for 100% patch branch coverage.npm run typechecknpm run test:coverage(local)Closes #2163
Made with Cursor