API Security

REST API Security Fundamentals

REST APIs are stateless by design: each request must contain all information needed for Authentication and authorization. The server does not maintain session state between requests.

Stateless Authentication Per Request

Every request must include:
1. Authentication credential (Bearer token, API key, mTLS certificate)
2. Required headers (Content-Type, Accept)
3. Any correlation/tracing identifiers

The server validates credentials on every request.
No server-side session is required (but tokens must be stateless or validated).

Authentication Methods

API Keys

# Simple API key in header
API_KEYS = {
    "key_abc123': {'name': "service-a'', "scopes': ['read:users']},
    'key_def456': {'name': "service-b'', "scopes': ['read:users', 'write:orders']},
}

@app.before_request
def validate_api_key():
    api_key = request.headers.get('X-API-Key')
    if not api_key or api_key not in API_KEYS:
        return jsonify({"error": "Invalid API key"}), 401
    g.api_key_info = API_KEYS[api_key]

API keys provide weak authentication: they are long-lived shared secrets that cannot be scoped to a User or rotated. Use them only for server-to-server communication where OAuth 2.0 is Impractical. Prefer OAuth 2.0 Bearer tokens for user-facing APIs.

Bearer Tokens

from functools import wraps

def require_bearer_token(required_scopes=None):
    def decorator(f):
        @wraps(f)
        def wrapper(*args, **kwargs):
            auth_header = request.headers.get('Authorization', '')
            if not auth_header.startswith('Bearer '):
                return jsonify({"error": "Missing Bearer token"}), 401

            token = auth_header[7:]
            try:
                claims = validate_jwt(token)
            except InvalidTokenError:
                return jsonify({"error": "Invalid or expired token"}), 401

            if required_scopes:
                token_scopes = set(claims.get('scope', '').split())
                missing = set(required_scopes) - token_scopes
                if missing:
                    return jsonify({"error": f"Missing scopes: {missing}"}), 403

            g.user = claims
            return f(*args, **kwargs)
        return wrapper
    return decorator

@app.route('/api/orders')
@require_bearer_token(['read:orders'])
def list_orders():
    ...

HMAC Signatures

import hmac
import hashlib
import time

def verify_hmac_signature(request, api_secret):
    timestamp = request.headers.get('X-Timestamp')
    signature = request.headers.get('X-Signature')

    if not timestamp or not signature:
        return False

    # Reject requests older than 5 minutes (replay protection)
    if abs(time.time() - int(timestamp)) > 300:
        return False

    # Recreate the signature
    message = f"{timestamp}.{request.method}.{request.path}.{request.get_data().decode()}"
    expected = hmac.new(
        api_secret.encode(),
        message.encode(),
        hashlib.sha256
    ).hexdigest()

    return hmac.compare_digest(signature, expected)

mTLS

# Nginx: require client certificates
server {
    listen 443 ssl;
    ssl_certificate /etc/nginx/server.crt;
    ssl_certificate_key /etc/nginx/server.key;
    ssl_client_certificate /etc/nginx/ca.crt;
    ssl_verify_client on;
    ssl_verify_depth 2;
}

Method	Security Level	Best For
API Key	Low	Internal services, prototyping
Bearer Token	High	User-facing APIs, microservices
HMAC	High	API gateways, financial APIs
mTLS	Very High	Service mesh, zero-trust networks

Authorization

RBAC (Role-Based Access Control)

# Role hierarchy
ROLES = {
    'viewer':   {'read'},
    'editor':   {'read', 'write'},
    'admin':    {'read', 'write', 'delete', 'admin'},
}

def require_role(role_name):
    def decorator(f):
        @wraps(f)
        def wrapper(*args, **kwargs):
            user_role = g.user.get('role', 'viewer')
            if user_role not in ROLES or role_name not in get_role_hierarchy(user_role):
                return jsonify({"error": "Insufficient permissions"}), 403
            return f(*args, **kwargs)
        return wrapper
    return decorator

ABAC (Attribute-Based Access Control)

def check_permission(user, resource, action):
    """Check if user can perform action on resource based on attributes."""
    if user['department'] == resource['department']:
        return True  # Same department: full access
    if user['role'] == 'admin':
        return True
    if action == 'read' and resource['visibility'] == 'public':
        return True
    return False

Scope-Based Authorization

@app.route('/api/admin/users')
@require_bearer_token(['admin:users'])
def admin_list_users():
    ...

@app.route('/api/orders')
@require_bearer_token(['read:orders'])
def list_orders():
    ...

Rate Limiting

Token Bucket Algorithm

import time
from collections import defaultdict

class TokenBucketRateLimiter:
    def __init__(self, capacity, refill_rate):
        self.capacity = capacity
        self.refill_rate = refill_rate  # tokens per second
        self.buckets = defaultdict(lambda: {'tokens': capacity, 'last_refill': time.time()})

    def allow(self, key):
        bucket = self.buckets[key]
        now = time.time()
        elapsed = now - bucket['last_refill']

        bucket['tokens'] = min(
            self.capacity,
            bucket['tokens'] + elapsed * self.refill_rate
        )
        bucket['last_refill'] = now

        if bucket['tokens'] >= 1:
            bucket['tokens'] -= 1
            return True
        return False

limiter = TokenBucketRateLimiter(capacity=100, refill_rate=10)

@app.before_request
def rate_limit():
    key = request.remote_addr  # or user ID, API key, etc.
    if not limiter.allow(key):
        return jsonify({"error": "Rate limit exceeded"}), 429

Sliding Window Log

import time
from collections import defaultdict

class SlidingWindowRateLimiter:
    def __init__(self, max_requests, window_seconds):
        self.max_requests = max_requests
        self.window_seconds = window_seconds
        self.requests = defaultdict(list)

    def allow(self, key):
        now = time.time()
        cutoff = now - self.window_seconds

        # Remove expired entries
        self.requests[key] = [t for t in self.requests[key] if t > cutoff]

        if len(self.requests[key]) >= self.max_requests:
            return False

        self.requests[key].append(now)
        return True

Rate Limiting Headers

X-RateLimit-Limit: 100
X-RateLimit-Remaining: 95
X-RateLimit-Reset: 1700000000
Retry-After: 30

Strategy	Per-User	Per-IP	Notes
Token bucket	Yes	Yes	Smooth rate, burst allowed
Sliding window	Yes	Yes	Precise, memory-intensive
Fixed window	Yes	Yes	Simple, burst at window boundaries

Rate limit by user identity (API key or token) when authenticated, and by IP address when Unauthenticated. Unauthenticated rate limits should be stricter to prevent abuse.

Input Validation

JSON Schema Validation

from jsonschema import validate, ValidationError

CREATE_ORDER_SCHEMA = {
    "type": "object",
    "required": ["customer_id", "items"],
    "properties": {
        "customer_id": {"type": "integer", "minimum": 1},
        "items": {
            "type": "array",
            "minItems": 1,
            "maxItems": 100,
            "items": {
                "type": "object",
                "required": ["product_id", "quantity"],
                "properties": {
                    "product_id": {"type": "integer", "minimum": 1},
                    "quantity": {"type": "integer", "minimum": 1, "maximum": 999}
                }
            }
        },
        "notes": {"type": "string", "maxLength": 1000}
    },
    "additionalProperties": False
}

@app.route('/api/orders', methods=['POST'])
@require_bearer_token(['write:orders'])
def create_order():
    data = request.get_json()
    try:
        validate(data, CREATE_ORDER_SCHEMA)
    except ValidationError as e:
        return jsonify({"error": f"Validation failed: {e.message}"}), 400
    ...

Type Checking and Length Limits

def validate_pagination(request):
    """Validate and sanitize pagination parameters."""
    try:
        limit = int(request.args.get('limit', 20))
        offset = int(request.args.get('offset', 0))
    except (ValueError, TypeError):
        return None, None, "limit and offset must be integers"

    limit = max(1, min(limit, 100))  # Clamp to [1, 100]
    offset = max(0, min(offset, 10000))  # Cap at 10000

    return limit, offset, None

CORS Configuration

from flask_cors import CORS

# SAFE: explicit allowlist
ALLOWED_ORIGINS = [
    'https://app.example.com',
    'https://admin.example.com'
]

CORS(app, origins=ALLOWED_ORIGINS, methods=['GET', 'POST', 'PUT', 'DELETE'],
     allow_headers=['Authorization', 'Content-Type'],
     supports_credentials=True)

# VULNERABLE: wildcard with credentials
# CORS(app, origins='*', supports_credentials=True)  # BLOCKED by browsers

API Versioning

Strategy	Example	Pros	Cons
URL path	`/api/v1/orders`	Simple, visible	URL changes
Header	`Accept: application/vnd.api.v1+json`	Clean URLs	Hidden, harder to test
Query parameter	`/api/orders?version=1`	Easy to add	Cache-busting issues

# URL path versioning (recommended for most APIs)
@app.route('/api/v1/orders')
def list_orders_v1():
    return jsonify(legacy_format(orders))

@app.route('/api/v2/orders')
def list_orders_v2():
    return jsonify(modern_format(orders))

Response Filtering

# Sparse fieldsets (JSON:API pattern)
@app.route('/api/v1/users')
def list_users():
    fields = request.args.get('fields', '').split(',')
    users = get_all_users()

    if fields:
        filtered = [{k: u[k] for k in fields if k in u} for u in users]
        return jsonify(filtered)
    return jsonify(users)

Pagination

Cursor vs Offset

# Offset pagination (simple, but slow at high offsets)
@app.route('/api/orders')
def list_orders():
    limit = min(int(request.args.get('limit', 20)), 100)
    offset = max(int(request.args.get('offset', 0)), 0)
    orders = db.query("SELECT * FROM orders ORDER BY id LIMIT %s OFFSET %s", (limit, offset))
    return jsonify({"data": orders, "offset": offset, "limit": limit})

# Cursor pagination (efficient at any position)
@app.route('/api/orders')
def list_orders():
    limit = min(int(request.args.get('limit', 20)), 100)
    cursor = request.args.get('cursor')

    if cursor:
        orders = db.query(
            "SELECT * FROM orders WHERE id > %s ORDER BY id LIMIT %s",
            (cursor, limit)
        )
    else:
        orders = db.query(
            "SELECT * FROM orders ORDER BY id LIMIT %s",
            (limit,)
        )

    next_cursor = orders[-1]['id'] if orders else None
    return jsonify({
        "data": orders,
        "next_cursor": next_cursor
    })

Method	Performance at page 10000	Consistency on inserts/deletes	URL bookmarkable
Offset	Slow (scans 10000+ rows)	Unstable (rows shift)	Yes
Cursor	Fast (index lookup)	Stable (deterministic)	No (cursor changes)

Idempotency Keys

from hashlib import sha256

@app.route('/api/orders', methods=['POST'])
@require_bearer_token(['write:orders'])
def create_order():
    idempotency_key = request.headers.get('Idempotency-Key')
    if not idempotency_key:
        return jsonify({"error": "Idempotency-Key header required"}), 400

    # Check if this key was already processed
    existing = redis.get(f"idempotency:{idempotency_key}")
    if existing:
        return jsonify(json.loads(existing)), 200  # Return original response

    # Process the order
    order = create_order_from_request(request)
    response = jsonify(order)

    # Store the response for future retries (TTL = 24h)
    redis.setex(
        f"idempotency:{idempotency_key}",
        86400,
        json.dumps(order)
    )

    return response, 201

Webhook Security

import hmac
import hashlib

def verify_webhook_signature(request, secret):
    signature = request.headers.get('X-Webhook-Signature')
    if not signature:
        return False

    payload = request.get_data()
    expected = hmac.new(
        secret.encode(),
        payload,
        hashlib.sha256
    ).hexdigest()

    return hmac.compare_digest(signature, expected)

@app.route('/webhooks/payment', methods=['POST'])
def payment_webhook():
    if not verify_webhook_signature(request, WEBHOOK_SECRET):
        return jsonify({"error": "Invalid signature"}), 401

    event = request.get_json()
    process_payment_event(event)
    return jsonify({"status": "ok"}), 200

API Gateway Patterns

API Gateway responsibilities:
1. Authentication (validate tokens, API keys)
2. Rate limiting (per-client, per-route)
3. Request transformation (headers, body)
4. Response transformation (filtering, formatting)
5. Load balancing (round-robin, least connections)
6. Circuit breaking (fail fast on downstream failures)
7. Request logging and monitoring
8. TLS termination

Gateway Implementation Options

Option	Complexity	Customization	Use Case
Kong	Medium	High (Lua plugins)	Large-scale, extensible
AWS API Gateway	Low	Medium	AWS ecosystem
Envoy	High	Very High	Service mesh, gRPC
Nginx	Medium	High	General purpose, lightweight
Traefik	Low	Medium	Container environments

GraphQL Security

Query Depth Limiting

# Limit maximum query depth to prevent complex/nested queries
MAX_QUERY_DEPTH = 5

def validate_query_depth(query_ast, current_depth=0):
    if current_depth > MAX_QUERY_DEPTH:
        raise GraphQLDepthError(f"Query depth exceeds maximum of {MAX_QUERY_DEPTH}")

    for field in query_ast.selection_set.selections:
        if hasattr(field, 'selection_set') and field.selection_set:
            validate_query_depth(field, current_depth + 1)

Complexity Analysis

# Assign cost to each field and limit total query complexity
FIELD_COSTS = {
    'orders': 10,
    'items': 5,
    'customer': 1,
}

MAX_COMPLEXITY = 500

def calculate_complexity(query_ast):
    total = 0
    for field in query_ast.selection_set.selections:
        field_name = field.name.value
        cost = FIELD_COSTS.get(field_name, 1)
        if hasattr(field, 'selection_set') and field.selection_set:
            cost *= calculate_complexity(field)
        total += cost
    return total

Introspection Control

// Disable introspection in production
const server = new ApolloServer({
  typeDefs,
  resolvers,
  introspection: process.env.NODE_ENV !== 'production',
});

OpenAPI Security

# openapi.yaml security schemes
components:
  securitySchemes:
    BearerAuth:
      type: http
      scheme: bearer
      bearerFormat: JWT

    ApiKeyAuth:
      type: apiKey
      in: header
      name: X-API-Key

    OAuth2:
      type: oauth2
      flows:
        authorizationCode:
          authorizationUrl: https://auth.example.com/authorize
          tokenUrl: https://auth.example.com/token
          scopes:
            read:orders: Read orders
            write:orders: Create/update orders

security:
  - BearerAuth: []
  - ApiKeyAuth: []

paths:
  /orders:
    get:
      security:
        - BearerAuth: [read:orders]
    post:
      security:
        - OAuth2: [write:orders]

Common Pitfalls

Not Validating Input on the Server

Client-side validation improves UX but provides zero security. An attacker can send any payload Directly to your API. Always validate on the server, regardless of what the client does.

CORS Misconfiguration

Reflecting the Origin header without validation allows any origin to make authenticated requests. Always use an explicit allowlist.

Exposing Internal IDs

Using sequential integer IDs (1, 2, 3…) allows attackers to enumerate resources. Use UUIDs or Hashids for public identifiers.

Not Rate Limiting Authentication Endpoints

Without rate limiting, attackers can brute-force credentials, enumerate usernames, and perform Credential stuffing attacks. Rate limit all authentication endpoints.

Returning Stack Traces in API Responses

Stack traces reveal implementation details (framework, library versions, file paths) that help Attackers craft targeted exploits. Return generic error messages in production; log details Server-side.

Summary

This topic covers the essential concepts and techniques related to api security, including key principles and practical applications.

Key concepts include:

core concepts and definitions
key principles and frameworks
practical applications
common techniques and methods
evaluation and critical analysis

A thorough understanding of these concepts, combined with regular practice and review, is essential for mastery of this topic.

Worked Examples

Worked examples demonstrating the application of key concepts are covered in the detailed sub-pages linked above.

Mark this page as reviewed