mcp-server-langgraph: Command for Claude Code

Generate Comprehensive Test Summary

Usage: /test-summary or /test-summary <scope>

Scopes:

• all - Run all tests and generate complete summary (default)
• unit - Unit tests only
• integration - Integration tests only
• quality - Property + contract + regression tests
• failed - Analyze failed tests only

---

🧪 Test Summary Generation

Step 1: Determine Test Scope

Based on $ARGUMENTS, determine which tests to run:

All Tests (default):

pytest -v --tb=short --cov=src --cov-report=term-missing

Unit Tests:

pytest -m unit -v --tb=short --cov=src --cov-report=term-missing

Integration Tests:

pytest -m integration -v --tb=short

Quality Tests:

pytest -m "property or contract or regression" -v --tb=short

---

Step 2: Run Tests and Collect Results

Execute tests and capture all output:

# Run tests with JSON output
pytest -v --tb=short --json-report --json-report-file=/tmp/test_results.json

# Also capture human-readable output
pytest -v --tb=short 2>&1 | tee /tmp/test_output.txt

Capture:

• Total tests discovered
• Tests passed/failed/skipped
• Test execution time
• Coverage percentage
• Failed test details (if any)

---

Step 3: Analyze Test Results

Parse test results from JSON and text output:

Summary Statistics:

{
    "total": <total tests>,
    "passed": <passed count>,
    "failed": <failed count>,
    "skipped": <skipped count>,
    "errors": <error count>,
    "pass_rate": <passed / total * 100>,
    "duration": <total seconds>,
    "coverage": <coverage percentage>
}

Failed Tests (if any):

• Test name and file location
• Failure reason (assertion, exception)
• Stack trace (abbreviated)
• Related code (file:line)

Skipped Tests:

• Test name
• Skip reason
• Category (infrastructure, experimental, etc.)

---

Step 4: Coverage Analysis

Analyze code coverage:

# Generate detailed coverage report
pytest --cov=src --cov-report=html --cov-report=term-missing

# Parse coverage summary
coverage report | grep "TOTAL"