Zero Script QA Skill

Quality assurance through logs and runtime verification, no test scripts needed.

What is Zero Script QA?

Zero Script QA verifies application quality by analyzing runtime behavior through Docker container logs and HTTP responses, eliminating the need to write traditional test scripts.

Actions

Action	Description	Example
`start`	Begin QA session	`$zero-script-qa start`
`verify`	Verify specific feature	`$zero-script-qa verify auth`
`report`	Generate QA report	`$zero-script-qa report`

Methodology

Step 1: Start Application with Docker

bash

docker compose up -d
docker compose logs -f app

Step 2: API Endpoint Verification

Test each endpoint with curl:

bash

# Health check
curl -s http://localhost:3000/api/health | jq .

# Create user
curl -s -X POST http://localhost:3000/api/users \
  -H "Content-Type: application/json" \
  -d '{"email":"test@test.com","name":"Test User","password":"pass123"}' | jq .

# Login
curl -s -X POST http://localhost:3000/api/auth/login \
  -H "Content-Type: application/json" \
  -d '{"email":"test@test.com","password":"pass123"}' | jq .

# Authenticated request
TOKEN=$(curl -s -X POST http://localhost:3000/api/auth/login \
  -H "Content-Type: application/json" \
  -d '{"email":"test@test.com","password":"pass123"}' | jq -r '.access_token')

curl -s http://localhost:3000/api/users/me \
  -H "Authorization: Bearer $TOKEN" | jq .

Step 3: Log Analysis

Check Docker logs for each operation:

bash

# Check for errors
docker compose logs app --tail 100 | grep -i error

# Check for warnings
docker compose logs app --tail 100 | grep -i warn

# Check database queries
docker compose logs db --tail 50

# Monitor real-time
docker compose logs -f --tail 0 app

Step 4: Edge Case Verification

bash

# Invalid input (expect 400)
curl -s -X POST http://localhost:3000/api/users \
  -H "Content-Type: application/json" \
  -d '{"email":"invalid","name":"","password":"short"}' | jq .

# Unauthorized access (expect 401)
curl -s http://localhost:3000/api/users/me | jq .

# Not found (expect 404)
curl -s http://localhost:3000/api/users/nonexistent-id | jq .

# Duplicate creation (expect 409)
curl -s -X POST http://localhost:3000/api/users \
  -H "Content-Type: application/json" \
  -d '{"email":"test@test.com","name":"Duplicate","password":"pass123"}' | jq .

Step 5: Performance Check

bash

# Response time measurement
time curl -s http://localhost:3000/api/health > /dev/null

# Multiple requests
for i in {1..10}; do
  time curl -s http://localhost:3000/api/users > /dev/null 2>&1
done

# Container resource usage
docker stats --no-stream

QA Report Template

markdown

# Zero Script QA Report

## Environment
- Date: YYYY-MM-DD
- Docker version: X.X.X
- Containers: app, db, redis

## Endpoint Tests

| Endpoint | Method | Expected | Actual | Status |
|----------|--------|----------|--------|--------|
| /api/health | GET | 200 | 200 | PASS |
| /api/users | POST | 201 | 201 | PASS |
| /api/auth/login | POST | 200 | 200 | PASS |
| /api/users (no auth) | GET | 401 | 401 | PASS |

## Error Log Summary
- Errors found: X
- Warnings found: X
- Details: [list each error]

## Performance
- Average response time: Xms
- Container memory: X MB
- Container CPU: X%

## Verdict: PASS / FAIL

QA Methodology Reference

See references/qa-methodology.md for detailed Zero Script QA patterns.

When to Use

•Rapid prototyping QA (no time for test scripts)
•API verification during development
•Integration testing after Phase 6
•Pre-deployment smoke testing
•BaaS application testing (bkend.ai)