🔧 GitHub Workflow Automation

Patterns for automating GitHub workflows with AI assistance, inspired by Gemini CLI and modern DevOps practices.

When to Use This Skill

Use this skill when:

• Automating PR reviews with AI
• Setting up issue triage automation
• Creating GitHub Actions workflows
• Integrating AI into CI/CD pipelines
• Automating Git operations (rebases, cherry-picks)

1. Automated PR Review

1.1 PR Review Action

# .github/workflows/ai-review.yml
name: AI Code Review

on:
  pull_request:
    types: [opened, synchronize]

jobs:
  review:
    runs-on: ubuntu-latest
    permissions:
      contents: read
      pull-requests: write

    steps:
      - uses: actions/checkout@v4
        with:
          fetch-depth: 0

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

      - name: Get diff
        id: diff
        run: |
          diff=$(git diff origin/${{ github.base_ref }}...HEAD)
          echo "diff<<EOF" >> $GITHUB_OUTPUT
          echo "$diff" >> $GITHUB_OUTPUT
          echo "EOF" >> $GITHUB_OUTPUT

      - name: AI Review
        uses: actions/github-script@v7
        with:
          script: |
            const { Anthropic } = require('@anthropic-ai/sdk');
            const client = new Anthropic({ apiKey: process.env.ANTHROPIC_API_KEY });

            const response = await client.messages.create({
              model: "claude-3-sonnet-20240229",
              max_tokens: 4096,
              messages: [{
                role: "user",
                content: `Review this PR diff and provide feedback:

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

                Diff:
                ${{ steps.diff.outputs.diff }}

                Provide:
                1. Summary of changes
                2. Potential issues or bugs
                3. Suggestions for improvement
                4. Security concerns if any

                Format as GitHub markdown.`
              }]
            });

            await github.rest.pulls.createReview({
              owner: context.repo.owner,
              repo: context.repo.repo,
              pull_number: context.issue.number,
              body: response.content[0].text,
              event: 'COMMENT'
            });
        env:
          ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}

1.2 Review Comment Patterns

# AI Review Structure

## 📋 Summary

Brief description of what this PR does.

## ✅ What looks good

- Well-structured code
- Good test coverage
- Clear naming conventions

## ⚠️ Potential Issues

1. **Line 42**: Possible null pointer exception
   ```javascript
   // Current
   user.profile.name;
   // Suggested
   user?.profile?.name ?? "Unknown";
   ```

1. Line 78: Consider error handling javascript // Add try-catch or .catch()

💡 Suggestions

• Consider extracting the validation logic into a separate function
• Add JSDoc comments for public methods

🔒 Security Notes

• No sensitive data exposure detected
• API key handling looks correct

### 1.3 Focused Reviews

```yaml
# Review only specific file types
- name: Filter code files
  run: |
    files=$(git diff --name-only origin/${{ github.base_ref }}...HEAD | \
            grep -E '\.(ts|tsx|js|jsx|py|go)$' || true)
    echo "code_files=$files" >> $GITHUB_OUTPUT

# Review with context
- name: AI Review with context
  run: |
    # Include relevant context files
    context=""
    for file in ${{ steps.changed.outputs.files }}; do
      if [[ -f "$file" ]]; then
        context+="=== $file ===\n$(cat $file)\n\n"
      fi
    done

    # Send to AI with full file context

2. Issue Triage Automation

2.1 Auto-label Issues

```yaml

.github/workflows/issue-triage.yml

name: Issue Triage

on: issues: types: [opened]

jobs: triage: runs-on: ubuntu-latest permissions: issues: write

steps:
  - name: Analyze issue
    uses: actions/github-script@v7
    with:
      script: |
        const issue = context.payload.issue;

        // Call AI to analyze
        const analysis = await analyzeIssue(issue.title, issue.body);

        // Apply labels
        const labels = [];

        if (analysis.type === 'bug') {
          labels.push('bug');
          if (analysis.severity === 'high') labels.push('priority: high');
        } else if (analysis.type === 'feature') {
          labels.push('enhancement');
        } else if (analysis.type === 'question') {
          labels.push('question');
        }

        if (analysis.area) {
          labels.push(`a