Pytest Runner Skill

Purpose

Single responsibility: Execute and manage pytest test suites with proper configuration, coverage reporting, and failure analysis. (BP-4)

Grounding Checkpoint (Archetype 1 Mitigation)

Before executing, VERIFY:

• Python virtual environment is active or available
• pytest is installed (pip show pytest)
• Test directory exists with test files
• pytest.ini or pyproject.toml configured (optional)

DO NOT run tests without verifying environment.

Uncertainty Escalation (Archetype 2 Mitigation)

ASK USER instead of guessing when:

• Multiple test directories detected - which to run?
• Coverage threshold unclear
• Parallel execution appropriate?
• Specific markers or keywords needed?

NEVER modify test configurations without user approval.

Context Scope (Archetype 3 Mitigation)

Context Type	Included	Excluded
RELEVANT	Test files, pytest config, fixtures	Application code details
PERIPHERAL	Coverage reports, test markers	CI/CD pipelines
DISTRACTOR	Other language tests	Deployment configs

Workflow Steps

Step 1: Environment Check (Grounding)

# Verify virtual environment
if [ -z "$VIRTUAL_ENV" ]; then
  # Activate if exists
  if [ -f "venv/bin/activate" ]; then
    source venv/bin/activate
  elif [ -f ".venv/bin/activate" ]; then
    source .venv/bin/activate
  else
    echo "WARNING: No virtual environment active"
  fi
fi

# Verify pytest installed
python -m pytest --version || pip install pytest

Step 2: Discover Tests

# List all test files
find . -name "test_*.py" -o -name "*_test.py" | head -20

# Show pytest collection
python -m pytest --collect-only -q

Step 3: Execute Tests

Basic execution:

python -m pytest tests/ -v

With coverage:

python -m pytest tests/ -v --cov=src --cov-report=term-missing --cov-report=html

Parallel execution:

python -m pytest tests/ -v -n auto  # requires pytest-xdist

With markers:

python -m pytest tests/ -v -m "unit"
python -m pytest tests/ -v -m "not slow"

Step 4: Analyze Results

# Parse test results
python -m pytest tests/ -v --tb=short 2>&1 | tee test_results.txt

# Extract failures
grep -E "^FAILED|^ERROR" test_results.txt

# Coverage summary
python -m pytest --cov=src --cov-report=term | grep -E "^TOTAL|^Name"

Recovery Protocol (Archetype 4 Mitigation)

On error:

1. PAUSE - Capture test output
2. DIAGNOSE - Check error type:
   • ImportError → Check dependencies, PYTHONPATH
   • FixtureError → Check conftest.py
   • CollectionError → Check test file syntax
   • Timeout → Reduce test scope or add markers
3. ADAPT - Adjust test selection or configuration
4. RETRY - With narrower scope (max 3 attempts)
5. ESCALATE - Report failures with context

Checkpoint Support

State saved to: .aiwg/working/checkpoints/pytest-runner/

checkpoints/pytest-runner/
├── test_collection.json     # Discovered tests
├── test_results.json        # Last run results
├── coverage_report.json     # Coverage data
└── failure_analysis.md      # Failure diagnostics

Common Pytest Options

Option	Purpose
-v	Verbose output
-x	Stop on first failure
-s	Show print statements
--lf	Run last failed tests
--ff	Run failed tests first
-k "pattern"	Filter by name pattern
-m "marker"	Filter by marker
--tb=short	Shorter tracebacks

Configuration Templates

pytest.ini:

[pytest]
testpaths = tests
python_files = test_*.py *_test.py
python_functions = test_*
addopts = -v --tb=short
markers =
    unit: Unit tests
    integration: Integration tests
    slow: Slow tests

pyproject.toml:

[tool.pytest.ini_options]
testpaths = ["tests"]
python_files = ["test_*.py", "*_test.py"]
addopts = "-v --tb=short"

References

• pytest documentation: https://docs.pytest.org/
• REF-001: Production-Grade Agentic Workflows (BP-4 single responsibility)
• REF-002: LLM Failure Modes (Archetype 1 grounding)