L’interface est localisée ; les présentations et documents des Skills sont traduits automatiquement par IA dans votre langue. Le code, les commandes, chemins et termes techniques restent en version originale, et le document de l’auteur fait foi.
Explorer les Skills
Déploiement et opérations · SKILL.md

workflow-setup

workflow-setup

Configures GitHub Actions CI/CD workflows for testing, linting, and deployment. Use when setting up automation for a Python, Rust, or TypeScript project.

Independent GitHub snapshot · Automated classification · Original documentation governs

Documentation de la Skill

GitHub

name: workflow-setup description: Configures GitHub Actions CI/CD workflows for testing, linting, and deployment. Use when setting up automation for a Python, Rust, or TypeScript project. globs: "**/.github/workflows/*.yml" alwaysApply: false

Custom metadata (not used by Claude for matching):

model: sonnet tools: [Read, Write, Bash] category: infrastructure tags: [github-actions, ci-cd, workflows, automation, testing] complexity: intermediate model_hint: standard estimated_tokens: 1500

Workflow Setup Skill

Set up GitHub Actions workflows for continuous integration and deployment.

When To Use

  • Need CI/CD for a new project
  • Adding missing workflows to existing project
  • Updating workflow versions to latest
  • Automating testing and quality checks
  • Setting up deployment pipelines

When NOT To Use

  • GitHub Actions workflows already configured and current
  • Project uses different CI platform (GitLab CI, CircleCI, etc.)
  • Not hosted on GitHub
  • Use /attune:upgrade-project instead for updating existing workflows

Standard Workflows

Python Workflows

  1. test.yml - Run pytest on push/PR
  2. lint.yml - Run ruff linting
  3. typecheck.yml - Run mypy type checking
  4. publish.yml - Publish to PyPI on release

Rust Workflows

  1. ci.yml - Combined test/lint/check workflow
  2. release.yml - Build and publish releases

TypeScript Workflows

  1. test.yml - Run Jest tests
  2. lint.yml - Run ESLint
  3. build.yml - Build for production
  4. deploy.yml - Deploy to hosting (Vercel, Netlify, etc.)

Workflow

1. Check Existing Workflows

ls -la .github/workflows/

Verification: Run the command with --help flag to verify availability.

2. Identify Missing Workflows

from project_detector import ProjectDetector

detector = ProjectDetector(Path.cwd())
language = detector.detect_language()

required_workflows = {
    "python": ["test.yml", "lint.yml", "typecheck.yml"],
    "rust": ["ci.yml"],
    "typescript": ["test.yml", "lint.yml", "build.yml"],
}

missing = detector.get_missing_configurations(language)

Verification: Run pytest -v to verify tests pass.

3. Render Workflow Templates

workflows_dir = Path(".github/workflows")
workflows_dir.mkdir(parents=True, exist_ok=True)

for workflow in required_workflows[language]:
    template = templates_dir / language / "workflows" / f"{workflow}.template"
    output = workflows_dir / workflow

    engine.render_file(template, output)
    print(f"✓ Created: {output}")

Verification: Run the command with --help flag to verify availability.

4. Validate Workflows

# Syntax check (requires act or gh CLI)
gh workflow list

# Or manually check YAML syntax
python3 -c "import yaml; yaml.safe_load(open('.github/workflows/test.yml'))"

Verification: Run pytest -v to verify tests pass.

Workflow Best Practices

Use Latest Action Versions

# Good - pinned to major version
- uses: actions/checkout@v4
- uses: actions/setup-python@v5

# Avoid - unpinned or outdated
- uses: actions/checkout@v2
- uses: actions/setup-python@latest

Verification: Run pytest -v to verify tests pass.

Matrix Testing (Python)

strategy:
  matrix:
    python-version: ["3.10", "3.11", "3.12"]
    os: [ubuntu-latest, macos-latest, windows-latest]

Verification: Run pytest -v to verify tests pass.

Caching Dependencies

- uses: actions/setup-python@v5
  with:
    python-version: '3.10'
    cache: 'pip'  # Cache pip dependencies

Verification: Run python --version to verify Python environment.

Shell Script Safety in Workflows

When writing inline shell scripts in workflows, ensure proper exit code handling:

# BAD - pipeline masks exit code
- run: |
    make typecheck 2>&1 | grep -v "^make\["
    echo "Typecheck passed"  # Runs even if make failed!

# GOOD - use pipefail
- run: |
    set -eo pipefail
    make typecheck 2>&1 | grep -v "^make\["

# GOOD - capture exit code explicitly
- run: |
    output=$(make typecheck 2>&1) || exit_code=$?
    echo "$output" | grep -v "^make\[" || true
    exit ${exit_code:-0}

For complex wrapper scripts, run /pensive:shell-review before integrating.

Updating Workflows

To update workflows to latest versions:

/attune:upgrade-project --component workflows

Verification: Run the command with --help flag to verify availability.

Related Skills

  • Skill(attune:project-init) - Full project initialization
  • Skill(sanctum:pr-prep) - PR preparation with CI checks

Exit Criteria

  • All required workflow files for the detected language exist under .github/workflows/ (Python: test.yml + lint.yml + typecheck.yml; Rust: ci.yml; TypeScript: test.yml + lint.yml + build.yml) and contain valid YAML syntax verified by python3 -c "import yaml; yaml.safe_load(open('...'))".
  • gh workflow list returns each created workflow file as an entry, confirming GitHub recognizes the workflow definitions.
  • Any inline shell script in a workflow uses set -eo pipefail or explicit exit-code capture; pipeline-masked failures (cmd | grep) without pipefail are flagged as errors.
  • If the project uses a CI platform other than GitHub Actions (GitLab CI, CircleCI), the skill reports this incompatibility and stops rather than generating GitHub-specific files.
Skills enregistrées