Files
ComfyUI_frontend/.github/workflows/claude-pr-review.yml
snomiao 080334754c fix: update Claude PR Review workflow to use correct action parameters (#5473)
* fix: update Claude PR Review workflow to use correct action parameters

- Changed 'direct_prompt' to 'prompt' (correct parameter name)
- Moved max_turns and timeout to claude_args parameter
- Changed allowed_tools to additional_permissions parameter

The workflow was failing silently because it was using invalid input parameters
that the claude-code-action doesn't recognize.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>

* fix: pin claude-code-action to v1.0.6 to prevent future breakage

Using @main tag could cause unexpected breakage when the action updates.
Pinning to a specific version ensures stability.

* fix: apply review feedback - correct migration to v1.0 format

- Moved timeout-minutes to job level (not in claude_args)
- Changed additional_permissions to --allowedTools in claude_args
- Fixed tool specification format per migration guide

These changes follow the official v0.x to v1.0 migration guide exactly.

---------

Co-authored-by: Claude <noreply@anthropic.com>
2025-09-09 23:25:36 -07:00

89 lines
3.1 KiB
YAML

name: Claude PR Review
permissions:
contents: read
pull-requests: write
issues: write
id-token: write
statuses: write
on:
pull_request:
types: [labeled]
jobs:
wait-for-ci:
runs-on: ubuntu-latest
if: github.event.label.name == 'claude-review'
outputs:
should-proceed: ${{ steps.check-status.outputs.proceed }}
steps:
- name: Wait for other CI checks
uses: lewagon/wait-on-check-action@e106e5c43e8ca1edea6383a39a01c5ca495fd812
with:
ref: ${{ github.event.pull_request.head.sha }}
check-regexp: '^(lint-and-format|test|playwright-tests)'
wait-interval: 30
repo-token: ${{ secrets.GITHUB_TOKEN }}
- name: Check if we should proceed
id: check-status
run: |
# Get all check runs for this commit
CHECK_RUNS=$(gh api repos/${{ github.repository }}/commits/${{ github.event.pull_request.head.sha }}/check-runs --jq '.check_runs[] | select(.name | test("lint-and-format|test|playwright-tests")) | {name, conclusion}')
# Check if any required checks failed
if echo "$CHECK_RUNS" | grep -q '"conclusion": "failure"'; then
echo "Some CI checks failed - skipping Claude review"
echo "proceed=false" >> $GITHUB_OUTPUT
else
echo "All CI checks passed - proceeding with Claude review"
echo "proceed=true" >> $GITHUB_OUTPUT
fi
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
claude-review:
needs: wait-for-ci
if: needs.wait-for-ci.outputs.should-proceed == 'true'
runs-on: ubuntu-latest
timeout-minutes: 30
steps:
- name: Checkout repository
uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Install pnpm
uses: pnpm/action-setup@v4
with:
version: 10
- name: Setup Node.js
uses: actions/setup-node@v4
with:
node-version: '20'
cache: 'pnpm'
- name: Install dependencies for analysis tools
run: |
pnpm install -g typescript @vue/compiler-sfc
- name: Run Claude PR Review
uses: anthropics/claude-code-action@v1.0.6
with:
label_trigger: "claude-review"
prompt: |
Read the file .claude/commands/comprehensive-pr-review.md and follow ALL the instructions exactly.
CRITICAL: You must post individual inline comments using the gh api commands shown in the file.
DO NOT create a summary comment.
Each issue must be posted as a separate inline comment on the specific line of code.
anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}
claude_args: "--max-turns 256 --allowedTools 'Bash(git:*),Bash(gh api:*),Bash(gh pr:*),Bash(gh repo:*),Bash(jq:*),Bash(echo:*),Read,Write,Edit,Glob,Grep,WebFetch'"
env:
PR_NUMBER: ${{ github.event.pull_request.number }}
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
COMMIT_SHA: ${{ github.event.pull_request.head.sha }}
BASE_SHA: ${{ github.event.pull_request.base.sha }}
REPOSITORY: ${{ github.repository }}