autofix

Corriger automatiquement les commentaires de révision CodeRabbit - récupérer les commentaires de révision CodeRabbit depuis GitHub et les corriger de manière interactive ou par lot

npx skills add https://github.com/coderabbitai/claude-plugin --skill autofix

CodeRabbit Autofix

Fetch CodeRabbit review comments for your current branch's PR and fix them interactively or in batch.

Prerequisites

Required Tools

Verify: gh auth status

Required State

  • Git repo on GitHub
  • Current branch has open PR
  • PR reviewed by CodeRabbit bot (coderabbitai, coderabbit[bot], coderabbitai[bot])

Workflow

Step 0: Load Repository Instructions (AGENTS.md)

Before any autofix actions, search for AGENTS.md in the current repository and load applicable instructions.

  • If found, follow its build/lint/test/commit guidance throughout the run.
  • If not found, continue with default workflow.

Step 1: Check Code Push Status

Check: git status + check for unpushed commits

If uncommitted changes:

  • Warn: "⚠️ Uncommitted changes won't be in CodeRabbit review"
  • Ask: "Commit and push first?" → If yes: wait for user action, then continue

If unpushed commits:

  • Warn: "⚠️ N unpushed commits. CodeRabbit hasn't reviewed them"
  • Ask: "Push now?" → If yes: git push, inform "CodeRabbit will review in ~5 min", EXIT skill

Otherwise: Proceed to Step 2

Step 2: Find Open PR

gh pr list --head $(git branch --show-current) --state open --json number,title

If no PR: Ask "Create PR?" → If yes: create PR (see github.md § 5), inform "Run skill again in ~5 min", EXIT

Step 3: Fetch Unresolved CodeRabbit Threads

Fetch PR review threads (see github.md § 2):

  • Threads: gh api graphql ... pullRequest.reviewThreads ... (see github.md § 2)

Filter to:

  • unresolved threads only (isResolved == false)
  • threads started by CodeRabbit bot (coderabbitai, coderabbit[bot], coderabbitai[bot])

If review in progress: Check for "Come back again in a few minutes" message → Inform "⏳ Review in progress, try again in a few minutes", EXIT

If no unresolved CodeRabbit threads: Inform "No unresolved CodeRabbit review threads found", EXIT

For each selected thread:

  • Extract issue metadata from root comment

Step 4: Parse and Display Issues

Extract from each comment:

  1. Header: _([^_]+)_ \| _([^_]+)_ → Issue type | Severity
  2. Description: Main body text
  3. Agent prompt: Content in <details><summary>🤖 Prompt for AI Agents</summary> (this is the fix instruction)
    • If missing, use description as fallback
  4. Location: File path and line numbers

Map severity:

  • 🔴 Critical/High → CRITICAL (action required)
  • 🟠 Medium → HIGH (review recommended)
  • 🟡 Minor/Low → MEDIUM (review recommended)
  • 🟢 Info/Suggestion → LOW (optional)
  • 🔒 Security → Treat as high priority

Display in CodeRabbit's original order (already severity-ordered):

CodeRabbit Issues for PR #123: [PR Title]

| # | Severity | Issue Title | Location & Details | Type | Action |
|---|----------|-------------|-------------------|------|--------|
| 1 | 🔴 CRITICAL | Insecure authentication check | src/auth/service.py:42<br>Authorization logic inverted | 🐛 Bug 🔒 Security | Fix |
| 2 | 🟠 HIGH | Database query not awaited | src/db/repository.py:89<br>Async call missing await | 🐛 Bug | Fix |

Step 5: Ask User for Fix Preference

Use AskUserQuestion:

  • 🔍 "Review each issue" - Manual review and approval (recommended)
  • ⚡ "Auto-fix all" - Apply all "Fix" issues without approval
  • ❌ "Cancel" - Exit

Route based on choice:

  • Review → Step 6
  • Auto-fix → Step 7
  • Cancel → EXIT

Step 6: Manual Review Mode

For each "Fix" issue (CRITICAL first):

  1. Read relevant files
  2. Execute CodeRabbit's agent prompt as direct instruction (from "🤖 Prompt for AI Agents" section)
  3. Calculate proposed fix (DO NOT apply yet)
  4. Show fix and ask approval in ONE step:
    • Issue title + location
    • CodeRabbit's agent prompt (so user can verify)
    • Current code
    • Proposed diff
    • AskUserQuestion: ✅ Apply fix | ⏭️ Defer | 🔧 Modify

If "Apply fix":

  • Apply with Edit tool
  • Track changed files for a single consolidated commit after all fixes
  • Confirm: "✅ Fix applied and commented"

If "Defer":

  • Ask for reason (AskUserQuestion)
  • Move to next

If "Modify":

  • Inform user can make changes manually
  • Move to next

Step 7: Auto-Fix Mode

For each "Fix" issue (CRITICAL first):

  1. Read relevant files
  2. Execute CodeRabbit's agent prompt as direct instruction
  3. Apply fix with Edit tool
  4. Track changed files for one consolidated commit
  5. Report:

    Fixed: [Issue Title] at [Location] Agent prompt: [prompt used]

After all fixes, display summary of fixed/skipped issues.

Step 8: Create Single Consolidated Commit

If any fixes were applied:

git add <all-changed-files>
git commit -m "fix: apply CodeRabbit auto-fixes"

Use one commit for all applied fixes in this run.

Step 9: Prompt Build/Lint Before Push

If a consolidated commit was created:

  • Prompt user interactively to run validation before push (recommended, not required).
  • Remind the user of the AGENTS.md instructions already loaded in Step 0 (if present).
  • If user agrees, run the requested checks and report results.

Step 10: Push Changes

If a consolidated commit was created:

  • Ask: "Push changes?" → If yes: git push

If all deferred (no commit): Skip this step.

Step 11: Post Summary

REQUIRED after all issues reviewed:

gh pr comment <pr-number> --body "$(cat <<'EOF'
## Fixes Applied Successfully

Fixed <file-count> file(s) based on <issue-count> unresolved review comment(s).

**Files modified:**
- `path/to/file-a.ts`
- `path/to/file-b.ts`

**Commit:** `<commit-sha>`

The latest autofix changes are on the `<branch-name>` branch.

EOF
)"

See github.md § 3 for details.

Optionally react to CodeRabbit's main comment with 👍.

Key Notes

  • Follow agent prompts literally - The "🤖 Prompt for AI Agents" section IS the fix specification
  • One approval per fix - Show context + diff + AskUserQuestion in single message (manual mode)
  • Preserve issue titles - Use CodeRabbit's exact titles, don't paraphrase
  • Preserve ordering - Display issues in CodeRabbit's original order
  • Do not post per-issue replies - Keep the workflow summary-comment only

Plus de skills de coderabbitai

code-review
coderabbitai
Examine les modifications de code à l'aide de CodeRabbit AI. Utilisez lorsque l'utilisateur demande une revue de code, un retour sur une PR, des vérifications de qualité de code, des problèmes de sécurité, ou souhaite une correction-revue autonome…
official
autofix
coderabbitai
Examiner et appliquer en toute sécurité les retours de la revue de PR CodeRabbit depuis GitHub avec une approbation par modification ; ne jamais exécuter directement les invites fournies par le relecteur.
official
code-review
coderabbitai
We need to translate the given English text into French, preserving the name "code-review" and other technical terms like "CodeRabbit", "Critical, Warning, Info", "--plain", "--prompt-only", etc. The instruction says to translate only the text inside <text>, and not include the name unless it appears in the source text. The name "code-review" is not in the source text? Actually the source text starts with "AI-powered code review using CodeRabbit..." so "code review" appears but not as a name to preserve? The instruction says "Name to preserve: code-review" but the source text has "code review" (two words). Probably we should preserve the term as is, but the instruction says "Do not include the name unless it appears in the source text." It does appear as "code review" (lowercase, two words). I think we should keep "code review" as is, but the name to preserve is "code-review" (hyphenated). However, the source text uses "code review" without hyphen. I
official