Smoke Test API Skill

Smoke Test API Skill

Quick Reference - Load this first for fast context (~3KB)

Mission

Execute comprehensive API health smoke tests to validate service availability, response times, and critical endpoint functionality during release workflows.

Core Capabilities

• Health Check Execution: Test API endpoints for availability and response time
• Critical Endpoint Validation: Verify core API operations (CRUD, search, auth)
• Response Validation: Check status codes, response structure, error handling
• Performance SLA Validation: Ensure response times meet performance budgets
• Multi-Environment Support: Test across pre-release, staging, production environments

When to Use This Skill

Use this skill when:

• Executing pre-release smoke tests (before staging deployment)
• Validating post-deployment health (after staging/production deployment)
• Testing canary deployments (5%, 25%, 100% traffic)
• Verifying rollback success (after rollback operations)
• Running scheduled health checks (monitoring integration)

Quick Start

1. Load Skill in Agent

skills:
  - name: smoke-test-api
    path: skills/smoke-test-api/SKILL.md

2. Execute Health Checks

const { SmokeTestAPI } = require('./scripts/execute-health-checks.js');

const tester = new SmokeTestAPI({
  baseUrl: 'https://staging.example.com',
  timeout: 5000,
  retries: 2
});

const result = await tester.executeHealthChecks({
  environment: 'staging',
  endpoints: [
    { path: '/health', method: 'GET', expectedStatus: 200 },
    { path: '/api/v1/users', method: 'GET', expectedStatus: 200 }
  ]
});

if (result.passed) {
  console.log('✅ API health checks passed');
} else {
  console.error('❌ API health checks failed');
}

3. Configuration Template

environment: staging
baseUrl: https://staging.example.com
timeout: 5000
retries: 2
endpoints:
  - name: health-check
    path: /health
    method: GET
    expectedStatus: 200
    sla: 100
  - name: list-users
    path: /api/v1/users
    method: GET
    expectedStatus: 200
    sla: 500

API Smoke Test Categories

1. Health Endpoints

• System health: /health, /ping, /status
• Database connectivity: /health/db
• Cache connectivity: /health/cache
• External service connectivity: /health/integrations

2. Critical CRUD Operations

• Create: POST /api/v1/resource (201 Created)
• Read: GET /api/v1/resource/:id (200 OK)
• Update: PUT /api/v1/resource/:id (200 OK)
• Delete: DELETE /api/v1/resource/:id (204 No Content)

3. Authentication & Authorization

• Login: POST /api/v1/auth/login (200 OK + token)
• Token validation: GET /api/v1/auth/me (200 OK with user)
• Protected resource: GET /api/v1/protected (401 without token, 200 with token)

4. Search & Query Operations

• Search: GET /api/v1/search?q=term (200 OK with results)
• Pagination: GET /api/v1/resource?page=1&limit=10 (200 OK with paginated results)
• Filtering: GET /api/v1/resource?status=active (200 OK with filtered results)

5. Error Handling

• Not found: GET /api/v1/nonexistent (404 Not Found)
• Bad request: POST /api/v1/resource with invalid data (400 Bad Request)
• Unauthorized: GET /api/v1/protected without token (401 Unauthorized)

Performance SLAs

const SLA_TARGETS = {
  health: 100,      // Health endpoints: ≤100ms
  read: 500,        // Read operations: ≤500ms
  write: 1000,      // Write operations: ≤1000ms
  search: 2000,     // Search operations: ≤2000ms
  timeout: 5000     // Global timeout: ≤5000ms
};

Pass/Fail Criteria

Pass: All smoke tests must pass

• ✅ All endpoints return expected status codes
• ✅ All response times meet SLA targets
• ✅ All response structures are valid
• ✅ No unexpected errors or exceptions

Fail: Any smoke test failure blocks deployment

• ❌ Any endpoint returns unexpected status code
• ❌ Any response time exceeds SLA target
• ❌ Any response structure is invalid
• ❌ Any unexpected error or exception occurs

Execution Points

API smoke tests are executed at these points in the release workflow:

1. Pre-Release (before staging deployment): Validate API readiness
2. Post-Staging (after staging deployment): Verify staging API health
3. Canary 5% (5% traffic): Test canary API with minimal traffic
4. Canary 25% (25% traffic): Test canary API with moderate traffic
5. Canary 100% (100% traffic): Test canary API with full traffic
6. Post-Production (after production deployment): Verify production API health
7. Post-Rollback (after rollback): Validate rollback API health

Common Patterns

Pattern 1: Simple Health Check

const result = await tester.testEndpoint({
  path: '/health',
  method: 'GET',
  expectedStatus: 200,
  sla: 100
});

Pattern 2: Authenticated Request

const result = await tester.testEndpoint({
  path: '/api/v1/users',
  method: 'GET',
  expectedStatus: 200,
  sla: 500,
  headers: {
    'Authorization': `Bearer ${token}`
  }
});

Pattern 3: POST with Body

const result = await tester.testEndpoint({
  path: '/api/v1/users',
  method: 'POST',
  expectedStatus: 201,
  sla: 1000,
  body: {
    name: 'Test User',
    email: 'test@example.com'
  }
});

Output Format

{
  passed: true,
  details: {
    totalEndpoints: 10,
    passed: 10,
    failed: 0,
    executionTime: 1234
  },
  reason: 'API smoke tests passed: All 10 endpoints responding',
  metrics: {
    endpointsPassed: 10,
    endpointsFailed: 0,
    averageResponseTime: 123,
    executionTime: 1234
  },
  results: [
    {
      endpoint: '/health',
      method: 'GET',
      status: 200,
      responseTime: 45,
      sla: 100,
      passed: true
    }
  ]
}

Need More Detail?

For comprehensive documentation including:

• Advanced testing patterns (retry logic, circuit breakers, rate limiting)
• Multi-environment configurations
• Integration with monitoring systems
• Custom validation rules
• Performance optimization strategies

Load: skills/smoke-test-api/REFERENCE.md (~15KB)