home / skills / jmagly / aiwg / tdd-enforce

This skill enforces test-first development by configuring pre-commit hooks and CI gates to ensure code cannot merge without adequate tests.

npx playbooks add skill jmagly/aiwg --skill tdd-enforce

Review the files below or copy the command above to add this skill to your agents.

Files (2)
SKILL.md
7.4 KB
---
name: tdd-enforce
description: Configure TDD enforcement via pre-commit hooks and CI coverage gates. Use when setting up test-first development workflow, adding coverage gates, or enforcing TDD practices.
version: 1.0.0
---

# TDD Enforce Skill

## Purpose

Configure Test-Driven Development enforcement through pre-commit hooks, CI coverage gates, and automated test execution. Ensures code cannot be committed or merged without adequate test coverage.

## Research Foundation

| Principle | Source | Reference |
|-----------|--------|-----------|
| TDD Methodology | Kent Beck (2002) | "Test-Driven Development by Example" |
| 80% Coverage | Google Testing Blog (2010) | [Code Coverage Goal](https://testing.googleblog.com/2010/07/code-coverage-goal-80-and-no-less.html) |
| Pre-commit Hooks | Industry Best Practice | [Husky](https://typicode.github.io/husky/), [pre-commit](https://pre-commit.com/) |
| CI Gates | ISTQB CT-TAS | [Test Automation Strategy](https://istqb.org/certifications/certified-tester-test-automation-strategy-ct-tas/) |

## When This Skill Applies

- User asks to "set up TDD" or "enforce test-first"
- User wants to "add coverage gates" or "block commits without tests"
- User mentions "pre-commit hooks for tests" or "CI test gates"
- Project needs test quality enforcement
- Brownfield project needs TDD adoption

## Trigger Phrases

| Natural Language | Action |
|------------------|--------|
| "Set up TDD enforcement" | Configure pre-commit + CI gates |
| "Add coverage gates" | Configure CI coverage thresholds |
| "Block commits without tests" | Set up pre-commit test hooks |
| "Enforce test-first development" | Full TDD setup |
| "Run tests on commit" | Configure pre-commit test execution |
| "Check if tests exist for new code" | Configure test presence validation |

## Configuration Options

### Coverage Thresholds

```yaml
coverage:
  line: 80        # Google standard: 80% minimum
  branch: 75      # Branch coverage threshold
  function: 90    # Function coverage threshold
  critical_paths: 100  # Auth, payments, validation
```

### Enforcement Levels

| Level | Pre-commit | CI Gate | Description |
|-------|-----------|---------|-------------|
| `strict` | Block | Fail | No exceptions, 100% enforcement |
| `standard` | Warn + Block | Fail | Standard TDD enforcement |
| `gradual` | Warn | Warn | For TDD adoption in brownfield |
| `audit` | Log only | Report | Visibility without blocking |

## Implementation Process

### 1. Detect Project Type

```python
def detect_project():
    """Identify project language and tooling"""
    if exists("package.json"):
        return "javascript"  # Use Husky
    elif exists("pyproject.toml") or exists("setup.py"):
        return "python"      # Use pre-commit
    elif exists("pom.xml") or exists("build.gradle"):
        return "java"        # Use maven/gradle hooks
    # ... etc
```

### 2. Install Pre-commit Hooks

**JavaScript (Husky + lint-staged)**:
```bash
npm install --save-dev husky lint-staged
npx husky init
```

`.husky/pre-commit`:
```bash
#!/usr/bin/env sh
. "$(dirname -- "$0")/_/husky.sh"

# Run tests for staged files
npx lint-staged

# Check coverage delta
npm run test:coverage -- --changedSince=HEAD~1
```

**Python (pre-commit)**:
```bash
pip install pre-commit
pre-commit install
```

`.pre-commit-config.yaml`:
```yaml
repos:
  - repo: local
    hooks:
      - id: pytest-check
        name: pytest-check
        entry: pytest --cov=src --cov-fail-under=80
        language: system
        types: [python]
        pass_filenames: false
```

### 3. Configure CI Coverage Gates

**GitHub Actions**:
```yaml
name: Test Coverage Gate
on: [pull_request]

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

      - name: Run tests with coverage
        run: npm test -- --coverage

      - name: Check coverage threshold
        run: |
          COVERAGE=$(jq '.total.lines.pct' coverage/coverage-summary.json)
          if (( $(echo "$COVERAGE < 80" | bc -l) )); then
            echo "Coverage $COVERAGE% is below 80% threshold"
            exit 1
          fi

      - name: Comment coverage on PR
        uses: actions/github-script@v7
        with:
          script: |
            const coverage = require('./coverage/coverage-summary.json');
            github.rest.issues.createComment({
              issue_number: context.issue.number,
              owner: context.repo.owner,
              repo: context.repo.repo,
              body: `## Coverage Report\n- Lines: ${coverage.total.lines.pct}%\n- Branches: ${coverage.total.branches.pct}%`
            });
```

### 4. Configure Test Presence Validation

Check that new/modified files have corresponding tests:

```python
def validate_test_presence(changed_files):
    """Ensure every source file has a test file"""
    missing_tests = []

    for src_file in changed_files:
        if is_source_file(src_file):
            test_file = get_test_file_path(src_file)
            if not exists(test_file):
                missing_tests.append(src_file)

    if missing_tests:
        print("ERROR: Missing tests for:")
        for f in missing_tests:
            print(f"  - {f}")
        return False
    return True
```

## Output Format

When reporting TDD enforcement setup:

```markdown
## TDD Enforcement Configured

### Pre-commit Hooks
- [x] Husky installed and initialized
- [x] Pre-commit hook: Run affected tests
- [x] Pre-commit hook: Check test presence for new files

### Coverage Gates
- Line coverage threshold: 80%
- Branch coverage threshold: 75%
- Critical path coverage: 100%

### CI Integration
- [x] GitHub Actions workflow created
- [x] Coverage gate: Fail PR if coverage < 80%
- [x] PR comment: Coverage report

### Files Created/Modified
- `.husky/pre-commit`
- `.github/workflows/test-coverage.yml`
- `package.json` (scripts updated)

### Next Steps
1. Run `npm test` to verify baseline coverage
2. Commit and push to test CI gates
3. Review coverage report on first PR
```

## Gradual Adoption for Brownfield

For existing projects without tests:

### Phase 1: Audit Mode (Week 1-2)
```yaml
enforcement: audit
# Log coverage but don't block
# Establish baseline
```

### Phase 2: New Code Only (Week 3-4)
```yaml
enforcement: gradual
coverage_delta: true  # Only check new code
threshold: 80
```

### Phase 3: Full Enforcement (Week 5+)
```yaml
enforcement: standard
threshold: 80
block_on_decrease: true
```

## Integration Points

- Works with `/setup-tdd` command
- Integrates with `mutation-test` skill for quality validation
- Feeds into `/flow-gate-check` for phase transitions
- Reports to Test Architect agent

## Related Skills

- `mutation-test` - Validate test quality beyond coverage
- `flaky-detect` - Identify unreliable tests
- `generate-factory` - Create test data infrastructure
- `test-sync` - Maintain test-code alignment

## Script Reference

### tdd_setup.py

Configure TDD enforcement for project:

```bash
# Standard enforcement (80% line, 75% branch)
python scripts/tdd_setup.py --level standard

# Custom thresholds
python scripts/tdd_setup.py --threshold 90 --branch-threshold 85

# Gradual adoption for brownfield projects
python scripts/tdd_setup.py --level gradual

# Preview without making changes
python scripts/tdd_setup.py --dry-run
```

Options:
- `--level`: Enforcement level (strict, standard, gradual, audit)
- `--threshold`: Line coverage threshold (default: 80)
- `--branch-threshold`: Branch coverage threshold (default: 75)
- `--dry-run`: Show what would be done without making changes

Overview

This skill configures Test-Driven Development enforcement using pre-commit hooks and CI coverage gates to prevent commits or merges without adequate tests. It automates test execution, coverage checks, and test-presence validation to embed TDD practices into developer workflows. Use it to adopt TDD from greenfield or to gradually enforce tests in brownfield projects.

How this skill works

The skill detects project type (JavaScript, Python, Java, etc.) and installs appropriate pre-commit tooling (e.g., Husky or pre-commit). It adds hooks that run affected tests and validate that changed source files have corresponding tests. In CI, it runs full test suites with coverage reporting and fails PRs that fall below configured thresholds, optionally commenting coverage summaries on the PR.

When to use it

  • Setting up test-first development for a new project
  • Adding CI coverage gates to block low-test changes
  • Enforcing test presence for changed source files
  • Introducing TDD gradually into an existing (brownfield) codebase
  • Preventing merges when coverage drops or critical paths lack tests

Best practices

  • Start in audit mode to establish a baseline before blocking changes
  • Use a gradual enforcement level for legacy code to avoid developer friction
  • Set realistic coverage thresholds (e.g., 80% lines, 75% branches) and tighten over time
  • Protect critical paths with 100% checks (auth, payments, validation)
  • Run only affected tests in pre-commit hooks to keep local feedback fast

Example use cases

  • Husky pre-commit hook runs changed tests and blocks commit if coverage delta is negative
  • GitHub Actions workflow fails PR if overall line coverage is below 80% and posts a coverage comment
  • Pre-commit script validates that every modified source file has a corresponding test file
  • Brownfield adoption: week 1 audit, week 3 new-code-only checks, week 5 full enforcement
  • Custom enforcement level: strict (block everything) for security-sensitive services

FAQ

Can I apply this to non-JavaScript projects?

Yes. The skill detects project type and uses appropriate tooling (pre-commit for Python, Maven/Gradle hooks for Java, etc.).

How do I avoid blocking work on large legacy codebases?

Use audit or gradual enforcement levels, enable coverage-delta checks for new code only, and raise thresholds over time.