fix: enforce safe prompt-perfector execution - #1439
Conversation
|
Warning Review limit reachedYou’ve reached a temporary PR review limit under our Fair Usage Limits Policy. Next review available in: 20 minutes Your organization has reached its usage spending cap. Adjust your spending cap in the billing tab. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (3)
📝 WalkthroughWalkthroughPrompt Perfector guidance now defines explicit prompt-handling and execution controls. Repository workflow documentation and a Git isolation verifier enforce state checks, while tests validate the safeguards and record the review outcome. ChangesPrompt Perfector safety workflow
Estimated code review effort: 3 (Moderate) | ~20 minutes Sequence Diagram(s)sequenceDiagram
participant Agent as Prompt Perfector agent
participant Verifier as verify-repository-isolation.mjs
participant Git as Git repository
Agent->>Verifier: request repository-state verification
Verifier->>Git: inspect worktree, branch, HEAD, operations, and status
Git-->>Verifier: repository state and status hash
Verifier-->>Agent: safety result, block reasons, and exit status
Possibly related PRs
Suggested labels: ✨ Finishing Touches📝 Generate docstrings
Comment |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 1611d5ee15
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
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".
|
@cursoragent resolve merge conflict |
|
Taking a look!   |
…ll-review-e608 Co-authored-by: BigSimmo <BigSimmo@users.noreply.github.com>
|
@cursoragent resolve merge conflict |
Resolve main sync fallout by keeping ledger history append-only, aligning skill contract tests with the workflow text, and documenting a POSIX bootstrap path when PowerShell is unavailable. Co-authored-by: BigSimmo <BigSimmo@users.noreply.github.com>
|
Closing as superseded by #1434. Current-head comparison confirms #1434 contains the same prompt-perfector SKILL.md and openai.yaml blobs, plus stronger repository-isolation workflow/verifier/tests (Codex Cloud support and content-sensitive dirty-state hashing). Keeping both would duplicate and potentially regress the stronger implementation. |
Pull request was closed
|
@cursoragent this PR will be re-opened soon. Please resolve any failing CI, run bugbot and resolve any bugbot comments against current repo and fix any merge conflicts and also fix any errors and address and fix all comments or suggestions that are likely to improve the PR and Codebase. Please implement smallest required fix according to best code practices. Check your work. Leave this PR closed for now but still make all these changes and run the review. I want it ready to go for when I re-open it. |
|
PR #1439 remains CLOSED and is ready to reopen on branch tip What I fixed
Review / Bugbot / CI
Checks run
Note for reopenThe branch tip is |
Keep main's Cloud/status-hash isolation verifier and restore an append-only ledger baseline, while preserving this branch's portable POSIX secondary worktree bootstrap and consolidated skill contract coverage. Co-authored-by: BigSimmo <BigSimmo@users.noreply.github.com>
Capture the post-main-sync review outcome for the closed prompt-perfector branch so it is ready when the PR is reopened. Co-authored-by: BigSimmo <BigSimmo@users.noreply.github.com>
Keep the closed PR branch one sync behind-clean before reopen. Co-authored-by: BigSimmo <BigSimmo@users.noreply.github.com>
|
@coderabbitai please run a new review of this closed PR for me |
|
✅ Action performedFull review finished. Your included review limit is currently reached under our Fair Usage Limits Policy. Your recent PR review activity is in the 95th percentile or higher among CodeRabbit users, so adaptive limits apply. This review may still proceed through usage-based billing if eligible. Your next included review will be available in 36 minutes. |
|
This pull request has been ignored for the connected project Preview Branches by Supabase. |
|
@cursoragent Work this open PR (confirm number + GitHub head first). Start from Goal: review + Bugbot the tip, fix actionable issues with smallest changes,
Stop when merge-tree clean, no actionable threads left, and required checks are |
|
PR #1439 · tip Snapshot → syncGitHub Review + BugbotUnique delta: POSIX secondary-worktree bootstrap, skill contract tests, one append-only ledger row (on main’s Cloud/status-hash verifier). No P0/P1; no new Bugbot findings; 0 unresolved threads. Restored ledger to append-only vs main. CI on tipSUCCESS: mergeability, policy, change scope, Gitleaks, Semgrep, Safety. In progress: Static PR checks, Unit coverage ( PostedSummary comment on the PR. Merge left to you. |
Unblock PR #1439: GitHub DIRTY was behind-but-clean vs advanced main. Co-authored-by: BigSimmo <BigSimmo@users.noreply.github.com>
Catch the one main commit that landed during the prior sync so GitHub can build refs/pull/1439/merge (local merge-tree already clean). Co-authored-by: BigSimmo <BigSimmo@users.noreply.github.com>
Restore docs/branch-review-ledger.md from origin/main and append only the current review+bugbot+fix record so the PR no longer reorders historical rows. Co-authored-by: BigSimmo <BigSimmo@users.noreply.github.com>
PR #1439 review+bugbot+fix summaryTip: Sync
Review + Bugbot
Fixed vs dispositioned
Required CI (tip
|
2 participants
Summary
prompt-perfectorrefinement-only by default with explicit evaluation, execution, and authority controls.Verification
npm run verify:pr-localnpm ci --include=devwas not run because Database Next.js and Vitest processes were active in other worktrees.npm run check:skills— 33 canonical skills and 8 aliases validquick_validate.py— skill validSAFE_TO_EDIT=trueBLOCK_REASON=primary_worktreenpm run check:branch-review-ledgergit diff --check origin/main...HEADUI verification not run: no UI, routing, styling, or browser behavior changed.
Risk and rollout
Clinical Governance Preflight
Not applicable: this PR does not touch ingestion, answer generation, search/ranking, source rendering, document access, privacy, production environment, or clinical output.
Notes
RAG impact: no retrieval behaviour change — prompt skill workflow and local repository-isolation tooling only.
The exact
npm run formatcommand could not start because local Prettier was absent. A cached offline Prettier 3.9.6 matching the lockfile formatted and checked all changed files; the whole-tree fallback was bounded but timed out without creating additional changes.Summary by CodeRabbit
New Features
Tests
Documentation