II.
Page JSON
Structured · livepage:docs-github-actions-setup-claude-code
Using Babysitter with Claude Code GitHub Actions json
Inspect the normalized record payload exactly as the atlas UI reads it.
{
"id": "page:docs-github-actions-setup-claude-code",
"_kind": "Page",
"_file": "wiki/docs/github-actions-setup-claude-code.md",
"_cluster": "wiki",
"attributes": {
"nodeKind": "Page",
"sourcePath": "docs/github-actions-setup-claude-code.md",
"sourceKind": "repo-docs",
"title": "Using Babysitter with Claude Code GitHub Actions",
"displayName": "Using Babysitter with Claude Code GitHub Actions",
"slug": "docs/github-actions-setup-claude-code",
"articlePath": "wiki/docs/github-actions-setup-claude-code.md",
"article": "\n# Using Babysitter with Claude Code GitHub Actions\n\nThis guide explains how to integrate the Babysitter plugin with [Claude Code GitHub Actions](https://github.com/anthropics/claude-code-action) for automated orchestration workflows in your CI/CD pipeline.\n\n## Overview\n\nThe Babysitter plugin enables deterministic, event-sourced workflow orchestration for Claude Code. When combined with GitHub Actions, you can automate complex multi-step development processes with quality gates, human approval checkpoints, and iterative refinement.\n\n## Quick Start\n\n### Basic Setup\n\nAdd the Babysitter plugin to your GitHub Actions workflow:\n\n```yaml\nname: Claude Code with Babysitter\n\non:\n issue_comment:\n types: [created]\n pull_request_review_comment:\n types: [created]\n issues:\n types: [opened, assigned]\n pull_request_review:\n types: [submitted]\n\njobs:\n claude:\n if: |\n (github.event_name == 'issue_comment' && contains(github.event.comment.body, '@claude')) ||\n (github.event_name == 'pull_request_review_comment' && contains(github.event.comment.body, '@claude')) ||\n (github.event_name == 'pull_request_review' && contains(github.event.review.body, '@claude')) ||\n (github.event_name == 'issues' && contains(github.event.issue.body, '@claude'))\n runs-on: ubuntu-latest\n permissions:\n contents: write\n pull-requests: write\n issues: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 1\n\n - name: Run Claude Code with Babysitter\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n\n # Add the Babysitter plugin marketplace\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n # Install the Babysitter plugin\n plugins: |\n babysitter@a5c.ai\n```\n\nFor comprehensive setup instructions, see the [Claude Code Action Setup Guide](https://github.com/anthropics/claude-code-action/blob/main/docs/setup.md).\n\n## Configuration Options\n\n### Plugin Marketplace\n\nThe Babysitter plugin is available from the a5c.ai marketplace:\n\n```yaml\nplugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n```\n\n### Plugin Installation\n\nInstall the plugin using its marketplace identifier:\n\n```yaml\nplugins: |\n babysitter@a5c.ai\n```\n\n## Workflow Examples\n\nSee the [official Claude Code Action examples](https://github.com/anthropics/claude-code-action/tree/main/examples) for additional workflow patterns.\n\n### Example 1: PR Review with Quality Gates\n\nUse Babysitter to orchestrate comprehensive PR reviews with progress tracking:\n\n```yaml\nname: Babysitter PR Review\n\non:\n pull_request:\n types: [opened, synchronize, ready_for_review, reopened]\n\njobs:\n review:\n runs-on: ubuntu-latest\n permissions:\n contents: read\n pull-requests: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 1\n\n - name: Run Babysitter PR Review\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call orchestrate a thorough code review using TDD Quality Convergence methodology.\n\n REPO: ${{ github.repository }}\n PR NUMBER: ${{ github.event.pull_request.number }}\n\n Analyze the PR changes for:\n - Code quality and best practices\n - Security vulnerabilities\n - Performance implications\n - Test coverage\n\n Generate a structured review report with:\n - Summary of changes\n - Issues found (critical, warning, info)\n - Recommendations\n```\n\n### Example 2: Feature Development with TDD\n\nAutomate test-driven development when issues are labeled:\n\n```yaml\nname: Babysitter TDD Feature\n\non:\n issues:\n types: [labeled]\n\njobs:\n develop:\n if: github.event.label.name == 'feature-request'\n runs-on: ubuntu-latest\n timeout-minutes: 30\n permissions:\n contents: write\n pull-requests: write\n issues: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 0\n\n - name: Run Babysitter TDD\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call implement the feature described in this issue using TDD Quality Convergence methodology.\n\n REPO: ${{ github.repository }}\n ISSUE NUMBER: ${{ github.event.issue.number }}\n\n The process should:\n 1. Write failing tests first\n 2. Implement minimal code to pass tests\n 3. Refactor for quality\n 4. Iterate until 80% quality threshold is met\n\n Create a PR when complete.\n```\n\n### Example 3: Spec-Driven Development\n\nFor projects with governance requirements using manual dispatch:\n\n```yaml\nname: Babysitter Spec-Kit\n\non:\n workflow_dispatch:\n inputs:\n spec_file:\n description: 'Path to specification file'\n required: true\n type: string\n\njobs:\n implement:\n runs-on: ubuntu-latest\n timeout-minutes: 60\n permissions:\n contents: write\n pull-requests: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 0\n\n - name: Run Babysitter Spec-Kit\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call implement the specification at ${{ inputs.spec_file }} using Spec-Kit methodology.\n\n Follow the spec-driven approach:\n 1. Parse and validate the specification\n 2. Generate implementation plan\n 3. Implement with continuous spec validation\n 4. Generate compliance report\n```\n\n### Example 4: GSD Quick Tasks\n\nFor rapid prototyping triggered by issue comments:\n\n```yaml\nname: Babysitter GSD\n\non:\n issue_comment:\n types: [created]\n\njobs:\n gsd:\n if: contains(github.event.comment.body, '/gsd')\n runs-on: ubuntu-latest\n timeout-minutes: 15\n permissions:\n contents: write\n pull-requests: write\n issues: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 1\n\n - name: Run Babysitter GSD\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n allowed_non_write_users: \"*\"\n github_token: ${{ secrets.GITHUB_TOKEN }}\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call use GSD methodology for the following task:\n\n ${{ github.event.comment.body }}\n\n Focus on rapid, working implementation with minimal overhead.\n```\n\n## Process Library Examples\n\nBabysitter includes a comprehensive process library with specialized processes for different domains. Use the `specializations/<domain>/<process>` syntax to invoke specific processes.\n\n### Available Specializations\n\n| Specialization | Description | Example Processes |\n|----------------|-------------|-------------------|\n| `devops-sre-platform` | DevOps, SRE, Platform Engineering | cicd-pipeline-setup, kubernetes-setup, incident-response |\n| `qa-testing-automation` | Testing and Quality Assurance | automation-framework, e2e-test-suite, performance-testing |\n| `security-compliance` | Security and Compliance | sast-pipeline, penetration-testing, gdpr-compliance |\n| `data-science-ml` | Data Science and Machine Learning | ml-project-scoping, model-training-pipeline, feature-engineering |\n| `data-engineering-analytics` | Data Engineering | etl-pipeline, data-warehouse, streaming-analytics |\n| `technical-documentation` | Documentation | adr-docs, runbook-docs, arch-docs-c4 |\n| `ux-ui-design` | UX/UI Design | design-system, accessibility-audit, usability-testing |\n| `software-architecture` | Architecture | microservices-design, api-design, system-design |\n\n### Example 5: CI/CD Pipeline Setup\n\nOrchestrate a complete CI/CD pipeline implementation:\n\n```yaml\nname: Babysitter CI/CD Setup\n\non:\n workflow_dispatch:\n inputs:\n target_environment:\n description: 'Target environment (dev, staging, prod)'\n required: true\n type: choice\n options:\n - dev\n - staging\n - prod\n\njobs:\n setup-cicd:\n runs-on: ubuntu-latest\n timeout-minutes: 45\n permissions:\n contents: write\n pull-requests: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 0\n\n - name: Run Babysitter CI/CD Pipeline Setup\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call use specializations/devops-sre-platform/cicd-pipeline-setup process.\n\n REPO: ${{ github.repository }}\n TARGET ENVIRONMENT: ${{ inputs.target_environment }}\n\n Set up a complete CI/CD pipeline with:\n - Build and test stages\n - Security scanning (SAST, dependency scanning)\n - Artifact management\n - Deployment to ${{ inputs.target_environment }}\n - Rollback capabilities\n```\n\n### Example 6: Test Automation Framework\n\nOrchestrate a comprehensive test automation framework setup:\n\n```yaml\nname: Babysitter Test Automation\n\non:\n issues:\n types: [labeled]\n\njobs:\n setup-automation:\n if: github.event.label.name == 'setup-testing'\n runs-on: ubuntu-latest\n timeout-minutes: 60\n permissions:\n contents: write\n pull-requests: write\n issues: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 0\n\n - name: Run Babysitter Test Automation Setup\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call use specializations/qa-testing-automation/automation-framework process.\n\n REPO: ${{ github.repository }}\n ISSUE NUMBER: ${{ github.event.issue.number }}\n\n Set up a test automation framework with:\n - Page Object Model architecture\n - Unit, integration, and E2E test structure\n - CI/CD pipeline integration\n - Test reporting and metrics\n - Cross-browser testing support\n\n Create a PR with the framework implementation.\n```\n\n### Example 7: Security Compliance (SAST Pipeline)\n\nAutomate security scanning pipeline setup:\n\n```yaml\nname: Babysitter Security Setup\n\non:\n workflow_dispatch:\n schedule:\n - cron: '0 0 * * 1' # Weekly on Monday\n\njobs:\n security-setup:\n runs-on: ubuntu-latest\n timeout-minutes: 30\n permissions:\n contents: write\n pull-requests: write\n security-events: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 0\n\n - name: Run Babysitter SAST Pipeline Setup\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call use specializations/security-compliance/sast-pipeline process.\n\n REPO: ${{ github.repository }}\n\n Implement a SAST (Static Application Security Testing) pipeline:\n - Code scanning with CodeQL or Semgrep\n - Dependency vulnerability scanning\n - Secret detection\n - Security findings reporting\n - Integration with GitHub Security tab\n```\n\n### Example 8: ML Model Training Pipeline\n\nOrchestrate ML model training workflows:\n\n```yaml\nname: Babysitter ML Training\n\non:\n workflow_dispatch:\n inputs:\n model_type:\n description: 'Model type'\n required: true\n type: string\n dataset_path:\n description: 'Path to training dataset'\n required: true\n type: string\n\njobs:\n ml-training:\n runs-on: ubuntu-latest\n timeout-minutes: 120\n permissions:\n contents: write\n pull-requests: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 0\n\n - name: Run Babysitter ML Training Pipeline\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call use specializations/data-science-ml/model-training-pipeline process.\n\n REPO: ${{ github.repository }}\n MODEL TYPE: ${{ inputs.model_type }}\n DATASET: ${{ inputs.dataset_path }}\n\n Implement a model training pipeline:\n - Data validation and preprocessing\n - Feature engineering\n - Model training with hyperparameter tuning\n - Model evaluation and metrics\n - Artifact storage and versioning\n - Experiment tracking\n```\n\n### Example 9: Incident Response Automation\n\nAutomate incident response procedures:\n\n```yaml\nname: Babysitter Incident Response\n\non:\n issues:\n types: [opened, labeled]\n\njobs:\n incident-response:\n if: contains(github.event.issue.labels.*.name, 'incident')\n runs-on: ubuntu-latest\n timeout-minutes: 30\n permissions:\n contents: read\n issues: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 1\n\n - name: Run Babysitter Incident Response\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call use specializations/devops-sre-platform/incident-response process.\n\n REPO: ${{ github.repository }}\n INCIDENT ISSUE: ${{ github.event.issue.number }}\n INCIDENT TITLE: ${{ github.event.issue.title }}\n\n Execute incident response procedure:\n - Analyze the incident description\n - Identify affected systems and severity\n - Generate investigation runbook\n - Create communication templates\n - Document timeline and action items\n```\n\n### Example 10: Architecture Documentation\n\nGenerate architecture documentation using C4 model:\n\n```yaml\nname: Babysitter Architecture Docs\n\non:\n workflow_dispatch:\n push:\n paths:\n - 'src/**'\n - 'packages/**'\n\njobs:\n generate-docs:\n runs-on: ubuntu-latest\n timeout-minutes: 30\n permissions:\n contents: write\n pull-requests: write\n id-token: write\n steps:\n - name: Checkout repository\n uses: actions/checkout@v6\n with:\n fetch-depth: 0\n\n - name: Run Babysitter Architecture Documentation\n uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n track_progress: true\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n prompt: |\n /babysitter:call use specializations/technical-documentation/arch-docs-c4 process.\n\n REPO: ${{ github.repository }}\n\n Generate architecture documentation:\n - C4 model diagrams (Context, Container, Component)\n - System overview and dependencies\n - Data flow documentation\n - API documentation\n - Update docs/ directory with generated content\n\n Create a PR with the documentation updates.\n```\n\n## Environment Variables\n\nConfigure Babysitter behavior through environment variables:\n\n```yaml\n- uses: anthropics/claude-code-action@v1\n with:\n anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }}\n\n plugin_marketplaces: |\n https://github.com/a5c-ai/babysitter.git\n\n plugins: |\n babysitter@a5c.ai\n\n env:\n # Keep runs under the checked-out repository for easy artifact upload\n BABYSITTER_RUNS_SCOPE: repo\n\n # Set maximum iterations (default: 65000)\n BABYSITTER_MAX_ITERATIONS: 100\n\n # Set quality threshold (default: 80)\n BABYSITTER_QUALITY_THRESHOLD: 85\n\n # Enable verbose logging\n BABYSITTER_LOG_LEVEL: debug\n```\n\n## Artifacts and Outputs\n\n### Preserving Run State\n\nSave Babysitter run artifacts for debugging:\n\n```yaml\n- uses: actions/upload-artifact@v4\n if: always()\n with:\n name: babysitter-runs\n path: .a5c/runs/\n retention-days: 7\n```\n\n### Run Outputs\n\nAccess run results in subsequent steps:\n\n```yaml\n- uses: anthropics/claude-code-action@v1\n id: babysitter\n with:\n # ... configuration ...\n\n- name: Check results\n run: |\n echo \"Run completed\"\n ls -la .a5c/runs/\n```\n\n## Best Practices\n\n### 1. Pin Plugin Versions\n\nFor reproducible builds, consider pinning to specific versions:\n\n```yaml\nplugins: |\n babysitter@a5c.ai@4.0.128\n```\n\n### 2. Set Reasonable Iteration Limits\n\nPrevent runaway workflows:\n\n```yaml\nenv:\n BABYSITTER_MAX_ITERATIONS: 50\n```\n\n### 3. Use Appropriate Methodologies\n\n| Use Case | Methodology |\n|----------|-------------|\n| Quick prototypes | GSD |\n| Quality-focused | TDD Quality Convergence |\n| Compliance required | Spec-Kit |\n| Team coordination | Scrum/Agile |\n\n### 4. Handle Failures Gracefully\n\n```yaml\n- uses: anthropics/claude-code-action@v1\n id: babysitter\n continue-on-error: true\n with:\n # ... configuration ...\n\n- name: Handle failure\n if: steps.babysitter.outcome == 'failure'\n run: |\n echo \"Babysitter run failed - check artifacts\"\n```\n\n## Troubleshooting\n\n### Plugin Not Loading\n\n1. Verify marketplace URL is accessible\n2. Check plugin name spelling\n3. Review action logs for installation errors\n\n### Runs Timing Out\n\n1. Reduce `BABYSITTER_MAX_ITERATIONS`\n2. Use simpler methodology (GSD vs TDD)\n3. Break task into smaller pieces\n\n### Permission Errors\n\nEnsure workflow has required permissions:\n\n```yaml\npermissions:\n contents: write # For commits\n pull-requests: write # For PR comments\n issues: write # For issue comments\n```\n\n## Related Documentation\n\n### Claude Code Action\n- [Setup Guide](https://github.com/anthropics/claude-code-action/blob/main/docs/setup.md) - Complete setup instructions\n- [Workflow Examples](https://github.com/anthropics/claude-code-action/tree/main/examples) - Official workflow examples\n- [Claude Code Action Repository](https://github.com/anthropics/claude-code-action)\n\n### Babysitter Plugin\n- [Getting Started](./reference/GETTING_STARTED.md)\n- [Process Selection Guide](https://github.com/a5c-ai/babysitter/blob/main/docs/reference/PROCESS_SELECTION.md)\n- [Troubleshooting](https://github.com/a5c-ai/babysitter/blob/main/docs/reference/TROUBLESHOOTING.md)\n",
"documents": []
},
"outgoingEdges": [],
"incomingEdges": []
}