Running Claude Code in GitHub Actions

Claude Code runs in GitHub Actions for automated code review, issue triage, and even automatic fixes.

Basic Setup

1. Store API key as secret

In your repository:

1. Settings → Secrets and variables → Actions
2. New repository secret
3. Name: ANTHROPIC_API_KEY
4. Value: your API key

2. Create workflow file

Create .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

      - 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 Review
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude --headless --auto-accept "Review the changes in this PR"

Common Workflows

PR Code Review

name: AI 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 dependencies
        run: npm install -g @anthropic-ai/claude-code

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

      - name: Review changes
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude --headless --auto-accept "
          Review these changed files for:
          1. Bugs or logic errors
          2. Security issues
          3. Performance concerns
          4. Code style issues

          Files: ${{ steps.changed.outputs.files }}
          " > review.txt

      - name: Post review comment
        uses: actions/github-script@v7
        with:
          script: |
            const fs = require('fs');
            const review = fs.readFileSync('review.txt', 'utf8');
            github.rest.issues.createComment({
              issue_number: context.issue.number,
              owner: context.repo.owner,
              repo: context.repo.repo,
              body: `## AI Code Review\n\n${review}`
            });

Auto-fix on PR

name: Auto-fix Issues

on:
  pull_request:
    types: [opened]

jobs:
  autofix:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
        with:
          ref: ${{ github.head_ref }}
          token: ${{ secrets.GITHUB_TOKEN }}

      - name: Setup
        run: npm install -g @anthropic-ai/claude-code

      - name: Fix issues
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude --headless --auto-accept "
          Fix any linting errors, type errors, or obvious bugs.
          Do not change functionality.
          "

      - name: Commit fixes
        run: |
          git config user.name "Claude Code"
          git config user.email "claude@anthropic.com"
          git add -A
          git diff --staged --quiet || git commit -m "fix: auto-fix issues"
          git push

Issue Triage

name: Triage Issues

on:
  issues:
    types: [opened]

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

      - name: Setup
        run: npm install -g @anthropic-ai/claude-code

      - name: Analyze issue
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        id: analysis
        run: |
          ISSUE_BODY="${{ github.event.issue.body }}"
          claude --headless "
          Analyze this issue and suggest:
          1. Appropriate labels (bug, feature, docs, etc.)
          2. Priority (high, medium, low)
          3. Which files might need changes

          Issue: $ISSUE_BODY
          " > analysis.txt

      - name: Add labels
        uses: actions/github-script@v7
        with:
          script: |
            const fs = require('fs');
            const analysis = fs.readFileSync('analysis.txt', 'utf8');

            // Parse labels from analysis
            const labels = [];
            if (analysis.includes('bug')) labels.push('bug');
            if (analysis.includes('feature')) labels.push('enhancement');
            if (analysis.includes('high priority')) labels.push('priority: high');

            if (labels.length > 0) {
              github.rest.issues.addLabels({
                issue_number: context.issue.number,
                owner: context.repo.owner,
                repo: context.repo.repo,
                labels: labels
              });
            }

Scheduled Maintenance

name: Weekly Maintenance

on:
  schedule:
    - cron: '0 0 * * 0'  # Every Sunday

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

      - name: Setup
        run: npm install -g @anthropic-ai/claude-code

      - name: Run maintenance
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
        run: |
          claude --headless --auto-accept "
          1. Update any outdated dependencies that are safe
          2. Remove unused imports
          3. Fix any new linting warnings
          "

      - name: Create PR
        uses: peter-evans/create-pull-request@v5
        with:
          title: "chore: weekly maintenance"
          branch: "maintenance/weekly"
          commit-message: "chore: weekly maintenance"

Cost Control

Limit when it runs

on:
  pull_request:
    types: [opened]  # Only on new PRs, not every push
    paths:
      - 'src/**'     # Only when source changes
      - '!**.md'     # Ignore docs

Use Haiku for simple tasks

- name: Quick check
  run: |
    claude --headless --model haiku "Check for obvious issues"

Cache results

- name: Cache review
  uses: actions/cache@v3
  with:
    path: .claude-cache
    key: review-${{ hashFiles('src/**') }}

- name: Review if not cached
  if: steps.cache.outputs.cache-hit != 'true'
  run: claude --headless "Review the code"

Security Best Practices

Limit permissions

permissions:
  contents: read
  pull-requests: write
  issues: write

Don’t expose secrets in logs

- name: Review
  env:
    ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
  run: |
    claude --headless "Review" 2>&1 | grep -v "api-key"

Validate before committing

- name: Validate changes
  run: |
    npm test
    npm run lint
    npm run typecheck

- name: Commit only if valid
  if: success()
  run: git commit -m "fix: auto-fix"

Debugging

Enable verbose output

- name: Debug run
  env:
    CLAUDE_DEBUG: "true"
  run: claude --headless "task"

Save outputs

- name: Review
  run: claude --headless "Review" > review.txt 2>&1

- name: Upload logs
  uses: actions/upload-artifact@v3
  with:
    name: claude-logs
    path: review.txt

Rate Limiting

Handle rate limits gracefully:

- name: Review with retry
  run: |
    for i in {1..3}; do
      claude --headless "Review" && break
      echo "Attempt $i failed, waiting..."
      sleep 60
    done

Quick Reference

Task	Trigger	Command
PR Review	pull_request	claude --headless "Review PR"
Auto-fix	pull_request	claude --headless --auto-accept "Fix issues"
Issue triage	issues:opened	claude --headless "Analyze issue"
Maintenance	schedule	claude --headless --auto-accept "Cleanup"