openapi: 3.0.3
info:
  title: My API
  version: 1.0.0
paths:
  /users:
    get:
      summary: List users
      parameters:
        - name: limit
          in: query
          schema:
            type: integer
            default: 20
      responses:
        '200':
          description: Success
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/UserList'
components:
  schemas:
    User:
      type: object
      required: [id, email]
      properties:
        id:
          type: string
        email:
          type: string
          format: email

GET /users?offset=40&limit=20

{
  "data": [...],
  "pagination": {
    "offset": 40,
    "limit": 20,
    "total": 150
  }
}

GET /users?cursor=abc123&limit=20

{
  "data": [...],
  "pagination": {
    "next_cursor": "def456",
    "has_more": true
  }
}

{
  "error": {
    "code": "VALIDATION_ERROR",
    "message": "Invalid request parameters",
    "details": [
      {
        "field": "email",
        "message": "Must be valid email format"
      }
    ],
    "request_id": "req_abc123"
  }
}

GET /users/123?fields=id,name,email

POST /users/bulk
{
  "operations": [
    { "method": "create", "data": {...} },
    { "method": "update", "id": "123", "data": {...} }
  ]
}

# Public, cacheable for 1 hour
Cache-Control: public, max-age=3600

# Private (user-specific), cacheable for 5 minutes
Cache-Control: private, max-age=300

# No caching at all
Cache-Control: no-store

# Cache but revalidate every time
Cache-Control: no-cache

# Stale content OK while revalidating
Cache-Control: max-age=60, stale-while-revalidate=30

# First request
GET /users/123
→ 200 OK
   ETag: "v1-abc123"
   Cache-Control: private, max-age=60

# Subsequent request (after cache expires)
GET /users/123
If-None-Match: "v1-abc123"
→ 304 Not Modified (cache still valid)
   OR
→ 200 OK with new ETag (content changed)

// Cache key design
const cacheKey = `${resource}:${id}:${version}`;

// Cache-aside pattern
async function getUser(id: string) {
  const cached = await cache.get(`user:${id}`);
  if (cached) return cached;

  const user = await db.users.findById(id);
  await cache.set(`user:${id}`, user, { ttl: 300 });
  return user;
}

// Cache invalidation on write
async function updateUser(id: string, data: UserUpdate) {
  const user = await db.users.update(id, data);
  await cache.delete(`user:${id}`);
  await cache.delete(`users:list:*`); // Invalidate list caches
  return user;
}

class TokenBucket {
  private tokens: number;
  private lastRefill: number;

  constructor(
    private capacity: number,      // Max burst size
    private refillRate: number,    // Tokens per second
  ) {
    this.tokens = capacity;
    this.lastRefill = Date.now();
  }

  consume(tokens: number = 1): boolean {
    this.refill();
    if (this.tokens >= tokens) {
      this.tokens -= tokens;
      return true;
    }
    return false;
  }

  private refill() {
    const now = Date.now();
    const elapsed = (now - this.lastRefill) / 1000;
    this.tokens = Math.min(
      this.capacity,
      this.tokens + elapsed * this.refillRate
    );
    this.lastRefill = now;
  }
}

HTTP/1.1 200 OK
X-RateLimit-Limit: 100
X-RateLimit-Remaining: 95
X-RateLimit-Reset: 1706792400

# When exceeded
HTTP/1.1 429 Too Many Requests
Retry-After: 30
X-RateLimit-Limit: 100
X-RateLimit-Remaining: 0
X-RateLimit-Reset: 1706792400

// Per-user limiting
const userKey = `ratelimit:user:${userId}`;

// Per-IP limiting (for anonymous)
const ipKey = `ratelimit:ip:${clientIp}`;

// Per-endpoint limiting
const endpointKey = `ratelimit:${method}:${path}`;

// Combined (most flexible)
const combinedKey = `ratelimit:${userId}:${method}:${path}`;

async function apiCallWithRetry(request: Request): Promise<Response> {
  const response = await fetch(request);

  if (response.status === 429) {
    const retryAfter = parseInt(
      response.headers.get('Retry-After') || '60'
    );

    console.log(`Rate limited. Retrying in ${retryAfter}s`);
    await sleep(retryAfter * 1000);
    return apiCallWithRetry(request);
  }

  return response;
}

// Proactive rate limiting
class RateLimitedClient {
  private remaining: number = Infinity;
  private resetTime: number = 0;

  async request(url: string): Promise<Response> {
    // Wait if we know we're out of quota
    if (this.remaining <= 0 && Date.now() < this.resetTime) {
      await sleep(this.resetTime - Date.now());
    }

    const response = await fetch(url);

    // Update tracking from headers
    this.remaining = parseInt(
      response.headers.get('X-RateLimit-Remaining') || '100'
    );
    this.resetTime = parseInt(
      response.headers.get('X-RateLimit-Reset') || '0'
    ) * 1000;

    return response;
  }
}

# AWS API Gateway
throttle:
  burstLimit: 100
  rateLimit: 50

# Azure API Management
<rate-limit calls="100" renewal-period="60" />

# Kong
plugins:
  - name: rate-limiting
    config:
      minute: 100
      policy: redis

□ Authentication on all non-public endpoints
□ Authorization checked for each resource
□ Rate limiting enabled
□ Input validation (size limits, type checking)
□ No sensitive data in URLs (tokens, passwords)
□ CORS configured appropriately
□ Security headers set
□ Audit logging for sensitive operations

□ Resource names are nouns, plural
□ HTTP methods match semantics
□ Status codes are appropriate
□ Error responses are consistent
□ Pagination is implemented for lists
□ Versioning strategy is clear
□ Authentication is documented
□ Rate limits are defined
□ Breaking changes are versioned
□ OpenAPI spec is accurate