Files
api/.github/workflows/claude-code-review.yml
renovate[bot] 541b0edd35 chore(deps): update actions/checkout action to v5 (#1581)
Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>
2025-08-18 13:30:54 -04:00

104 lines
4.6 KiB
YAML

name: Claude Code Review
on:
pull_request:
types: [opened, synchronize]
# Skip reviews for non-code changes
paths-ignore:
- "**/*.md"
- "**/package-lock.json"
- "**/pnpm-lock.yaml"
- "**/.gitignore"
- "**/LICENSE"
- "**/*.config.js"
- "**/*.config.ts"
- "**/tsconfig.json"
- "**/.github/workflows/*.yml"
- "**/docs/**"
jobs:
claude-review:
# Skip review for bot PRs and WIP/skip-review PRs
# Only run if changes are significant (>10 lines)
if: |
(github.event.pull_request.additions > 10 || github.event.pull_request.deletions > 10) &&
!contains(github.event.pull_request.title, '[skip-review]') &&
!contains(github.event.pull_request.title, '[WIP]') &&
!endsWith(github.event.pull_request.user.login, '[bot]') &&
github.event.pull_request.user.login != 'dependabot' &&
github.event.pull_request.user.login != 'renovate'
runs-on: ubuntu-latest
permissions:
contents: read
pull-requests: read
issues: read
id-token: write
steps:
- name: Checkout repository
uses: actions/checkout@v5
with:
fetch-depth: 1
- name: Run Claude Code Review
id: claude-review
uses: anthropics/claude-code-action@beta
with:
claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}
# Optional: Specify model (defaults to Claude Sonnet 4, uncomment for Claude Opus 4)
# model: "claude-opus-4-20250514"
# Direct prompt for automated review (no @claude mention needed)
direct_prompt: |
IMPORTANT: Review ONLY the DIFF/CHANGESET - the actual lines that were added or modified in this PR.
DO NOT review the entire file context, only analyze the specific changes being made.
Look for HIGH-PRIORITY issues in the CHANGED LINES ONLY:
1. CRITICAL BUGS: Logic errors, null pointer issues, infinite loops, race conditions
2. SECURITY: SQL injection, XSS, authentication bypass, exposed secrets, unsafe operations
3. BREAKING CHANGES: API contract violations, removed exports, changed function signatures
4. DATA LOSS RISKS: Destructive operations without safeguards, missing data validation
DO NOT comment on:
- Code that wasn't changed in this PR
- Style, formatting, or documentation
- Test coverage (unless tests are broken by the changes)
- Minor optimizations or best practices
- Existing code issues that weren't introduced by this PR
If you find no critical issues in the DIFF, respond with: "✅ No critical issues found in changes"
Keep response under 10 lines. Reference specific line numbers from the diff when reporting issues.
# Optional: Use sticky comments to make Claude reuse the same comment on subsequent pushes to the same PR
use_sticky_comment: true
# Context-aware review based on PR characteristics
# Uncomment to enable different review strategies based on context
# direct_prompt: |
# ${{
# (github.event.pull_request.additions > 500) &&
# 'Large PR detected. Focus only on architectural issues and breaking changes. Skip minor issues.' ||
# contains(github.event.pull_request.title, 'fix') &&
# 'Bug fix PR: Verify the fix addresses the root cause and check for regression risks.' ||
# contains(github.event.pull_request.title, 'deps') &&
# 'Dependency update: Check for breaking changes and security advisories only.' ||
# contains(github.event.pull_request.title, 'refactor') &&
# 'Refactor PR: Verify no behavior changes and check for performance regressions.' ||
# contains(github.event.pull_request.title, 'feat') &&
# 'New feature: Check for security issues, edge cases, and integration problems only.' ||
# 'Standard review: Check for critical bugs, security issues, and breaking changes only.'
# }}
# Optional: Add specific tools for running tests or linting
# allowed_tools: "Bash(npm run test),Bash(npm run lint),Bash(npm run typecheck)"
# Optional: Skip review for certain conditions
# if: |
# !contains(github.event.pull_request.title, '[skip-review]') &&
# !contains(github.event.pull_request.title, '[WIP]')