GitHub Actions Integration

1 min read

Automating workflows with Claude Code in CI/CD pipelines

title: GitHub Actions Integration description: Automating workflows with Claude Code in CI/CD pipelines

Integrate Claude Code into your GitHub Actions workflows for automated code review, documentation generation, test creation, and more.

Basic Setup

Prerequisites

Claude Code installed in your CI environment
ANTHROPIC_API_KEY stored as a GitHub secret
A GitHub repository with Actions enabled

Simple Workflow

YAML

# .github/workflows/claude-review.yml
name: Claude Code Review

on:
  pull_request:
    types: [opened, synchronize]

jobs:
  review:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Setup Node.js
        uses: actions/setup-node@v4
        with:
          node-version: '20'

      - name: Install Claude Code
        run: npm install -g @anthropic-ai/claude-code

      - name: Run Code Review
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude "Review the changes in this PR and provide feedback" \
            --output-format json > review.json

      - name: Post Review Comment
        uses: actions/github-script@v7
        with:
          script: |
            const fs = require('fs');
            const review = JSON.parse(fs.readFileSync('review.json', 'utf8'));
            github.rest.issues.createComment({
              owner: context.repo.owner,
              repo: context.repo.repo,
              issue_number: context.issue.number,
              body: review.content
            });

Common Use Cases

Automated Code Review

YAML

name: AI Code Review

on:
  pull_request:
    paths:
      - '**.ts'
      - '**.tsx'
      - '**.js'
      - '**.jsx'

jobs:
  review:
    runs-on: ubuntu-latest
    permissions:
      pull-requests: write
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0

      - name: Get changed files
        id: changed
        run: |
          echo "files=$(git diff --name-only origin/${{ github.base_ref }}...HEAD | tr '\n' ' ')" >> $GITHUB_OUTPUT

      - name: Install Claude Code
        run: npm install -g @anthropic-ai/claude-code

      - name: Review changes
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude "Review these changed files for:
          - Code quality issues
          - Potential bugs
          - Security concerns
          - Performance problems

          Files: ${{ steps.changed.outputs.files }}

          Format your response as markdown with sections for each category."

Documentation Generation

YAML

name: Generate Docs

on:
  push:
    branches: [main]
    paths:
      - 'src/**/*.ts'

jobs:
  docs:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Install Claude Code
        run: npm install -g @anthropic-ai/claude-code

      - name: Generate documentation
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude "Generate API documentation for all exported functions in src/.
          Create markdown files in docs/ directory.
          Include parameter descriptions, return types, and usage examples."

      - name: Commit docs
        run: |
          git config user.name "github-actions[bot]"
          git config user.email "github-actions[bot]@users.noreply.github.com"
          git add docs/
          git diff --staged --quiet || git commit -m "docs: auto-generated documentation"
          git push

Test Generation

YAML

name: Generate Tests

on:
  pull_request:
    paths:
      - 'src/**/*.ts'
      - '!src/**/*.test.ts'

jobs:
  tests:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Install dependencies
        run: |
          npm ci
          npm install -g @anthropic-ai/claude-code

      - name: Find untested files
        id: untested
        run: |
          # Find source files without corresponding test files
          untested=""
          for file in src/**/*.ts; do
            test_file="${file%.ts}.test.ts"
            if [ ! -f "$test_file" ]; then
              untested="$untested $file"
            fi
          done
          echo "files=$untested" >> $GITHUB_OUTPUT

      - name: Generate tests
        if: steps.untested.outputs.files != ''
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude "Generate unit tests for these files: ${{ steps.untested.outputs.files }}

          Requirements:
          - Use Jest and React Testing Library
          - Cover edge cases
          - Mock external dependencies
          - Follow existing test patterns in the codebase"

      - name: Run tests
        run: npm test

Commit Message Enhancement

YAML

name: Enhance Commit Messages

on:
  push:
    branches: [main]

jobs:
  enhance:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 2

      - name: Install Claude Code
        run: npm install -g @anthropic-ai/claude-code

      - name: Analyze commit
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          COMMIT_MSG=$(git log -1 --pretty=%B)
          DIFF=$(git diff HEAD~1)

          claude "Given this commit message and diff, suggest an improved commit message that:
          - Follows conventional commits format
          - Is more descriptive
          - Includes the scope

          Original message: $COMMIT_MSG
          Diff: $DIFF"

Advanced Patterns

Conditional Reviews

YAML

name: Smart Review

on:
  pull_request:

jobs:
  analyze:
    runs-on: ubuntu-latest
    outputs:
      needs-review: ${{ steps.check.outputs.needs-review }}
      complexity: ${{ steps.check.outputs.complexity }}
    steps:
      - uses: actions/checkout@v4

      - name: Check PR complexity
        id: check
        run: |
          LINES_CHANGED=$(git diff --shortstat origin/${{ github.base_ref }} | awk '{print $4+$6}')
          FILES_CHANGED=$(git diff --name-only origin/${{ github.base_ref }} | wc -l)

          if [ "$LINES_CHANGED" -gt 100 ] || [ "$FILES_CHANGED" -gt 5 ]; then
            echo "needs-review=true" >> $GITHUB_OUTPUT
            echo "complexity=high" >> $GITHUB_OUTPUT
          else
            echo "needs-review=false" >> $GITHUB_OUTPUT
            echo "complexity=low" >> $GITHUB_OUTPUT
          fi

  review:
    needs: analyze
    if: needs.analyze.outputs.needs-review == 'true'
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Install Claude Code
        run: npm install -g @anthropic-ai/claude-code

      - name: Deep review
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude "This is a ${{ needs.analyze.outputs.complexity }} complexity PR.
          Perform a thorough review focusing on:
          - Architecture decisions
          - Breaking changes
          - Test coverage
          - Documentation needs"

Multi-Model Strategy

YAML

name: Multi-Model Review

jobs:
  quick-check:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Quick lint check
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
          ANTHROPIC_MODEL: claude-3-5-haiku-20241022
        run: |
          claude "Quick check: any obvious issues in the changed files?"

  deep-review:
    needs: quick-check
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Deep analysis
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
          ANTHROPIC_MODEL: claude-opus-4-20250514
        run: |
          claude "Perform comprehensive code review..."

Caching for Performance

YAML

name: Cached Review

jobs:
  review:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4

      - name: Cache Claude context
        uses: actions/cache@v4
        with:
          path: ~/.claude
          key: claude-${{ github.repository }}-${{ hashFiles('CLAUDE.md') }}

      - name: Install Claude Code
        run: npm install -g @anthropic-ai/claude-code

      - name: Review with context
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: claude "Review the PR changes"

Security Considerations

Secrets Management

YAML

# Never log or expose the API key
- name: Safe usage
  env:
    ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
  run: |
    # Don't echo the key
    claude "Review code"

Permission Scoping

YAML

# Minimal permissions
permissions:
  contents: read
  pull-requests: write  # Only if commenting

jobs:
  review:
    runs-on: ubuntu-latest
    # ...

Input Sanitization

YAML

- name: Sanitize PR title
  run: |
    # Don't pass untrusted input directly
    SAFE_TITLE=$(echo "${{ github.event.pull_request.title }}" | tr -cd '[:alnum:] ')
    claude "Review PR: $SAFE_TITLE"

Troubleshooting

Rate Limits

YAML

- name: Handle rate limits
  env:
    ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
  run: |
    for i in 1 2 3; do
      claude "Review code" && break
      echo "Rate limited, waiting..."
      sleep 60
    done

Large Diffs

YAML

- name: Handle large PRs
  run: |
    DIFF_SIZE=$(git diff --shortstat origin/${{ github.base_ref }} | awk '{print $4+$6}')

    if [ "$DIFF_SIZE" -gt 1000 ]; then
      # Review file by file for large PRs
      for file in $(git diff --name-only origin/${{ github.base_ref }}); do
        claude "Review $file"
      done
    else
      claude "Review all changes"
    fi

Timeout Handling

YAML

- name: Review with timeout
  timeout-minutes: 10
  env:
    ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
  run: |
    timeout 300 claude "Review code" || echo "Review timed out"

Best Practices

Use secrets - Never hardcode API keys
Scope permissions - Request minimal GitHub permissions
Handle failures gracefully - Don't block PRs on review failures
Cache when possible - Reduce API calls and costs
Use appropriate models - Haiku for quick checks, Opus for deep reviews
Log responsibly - Don't log sensitive information
Test workflows - Use act for local testing

Next Steps

Sources

Generated with AI using Claude AI by Anthropic

Model: Claude Opus 4.5 · Generated: 2025-12-09 · Build: v0.9.0-b4563d6

Edit this page on GitHub