Securing Auth Routes

Goal

Harden all authentication-related endpoints against common attacks by implementing CSRF protection, rate limiting, HTTPS enforcement, security headers, and brute force mitigation.

When to Use

•After implementing auth routes (login, registration, password reset, OAuth callbacks).
•When preparing an application for production deployment.
•During a security review or hardening pass.
•When adding new auth-related endpoints to an existing application.

Instructions

CSRF Protection

For browser-based forms that are not purely API-driven, use SameSite cookies combined with a CSRF token.

python

import secrets
from flask import session, request, abort

def generate_csrf_token() -> str:
    """Generate a CSRF token and store it in the session."""
    if "csrf_token" not in session:
        session["csrf_token"] = secrets.token_urlsafe(32)
    return session["csrf_token"]


def validate_csrf_token():
    """Validate the CSRF token from the request against the session."""
    token = request.form.get("csrf_token") or request.headers.get("X-CSRF-Token")
    if not token or token != session.get("csrf_token"):
        abort(403, description="CSRF validation failed")

For API endpoints that use Authorization: Bearer headers (not cookies), CSRF protection is typically unnecessary because the token must be explicitly attached by the client. SameSite=Strict cookies provide an additional layer of protection for cookie-based auth.

Rate Limiting Auth Endpoints

Apply strict rate limits to all authentication endpoints to prevent brute force and credential stuffing attacks.

python

from functools import wraps
from datetime import datetime, timedelta, timezone

# In-memory store for demonstration; use Redis in production
rate_limit_store: dict[str, list[datetime]] = {}


def rate_limit(max_requests: int, window: timedelta):
    """Decorator to rate limit an endpoint by client IP."""
    def decorator(f):
        @wraps(f)
        def wrapped(*args, **kwargs):
            ip = request.remote_addr
            key = f"{f.__name__}:{ip}"
            now = datetime.now(timezone.utc)

            # Clean old entries
            entries = rate_limit_store.get(key, [])
            entries = [t for t in entries if now - t < window]

            if len(entries) >= max_requests:
                return jsonify({"error": "Too many requests. Try again later."}), 429

            entries.append(now)
            rate_limit_store[key] = entries
            return f(*args, **kwargs)
        return wrapped
    return decorator

Recommended limits:

Endpoint	Limit
Login	5 attempts per minute per IP
Registration	3 attempts per hour per IP
Password reset	3 attempts per hour per IP
Token refresh	10 per minute per IP

python

@auth_bp.route("/login", methods=["POST"])
@rate_limit(max_requests=5, window=timedelta(minutes=1))
def login():
    # ... login logic
    pass


@auth_bp.route("/register", methods=["POST"])
@rate_limit(max_requests=3, window=timedelta(hours=1))
def register():
    # ... registration logic
    pass

HTTPS Enforcement

Redirect all HTTP requests to HTTPS and set the HSTS header to instruct browsers to always use HTTPS.

python

from flask import redirect, request

@app.before_request
def enforce_https():
    if not request.is_secure and not app.debug:
        url = request.url.replace("http://", "https://", 1)
        return redirect(url, code=301)


@app.after_request
def set_security_headers(response):
    response.headers["Strict-Transport-Security"] = "max-age=31536000; includeSubDomains"
    response.headers["X-Content-Type-Options"] = "nosniff"
    response.headers["X-Frame-Options"] = "DENY"
    response.headers["Referrer-Policy"] = "strict-origin-when-cross-origin"
    return response

Security Headers

Apply these headers to every response:

Header	Value	Purpose
Strict-Transport-Security	`max-age=31536000; includeSubDomains`	Enforce HTTPS for 1 year
X-Content-Type-Options	`nosniff`	Prevent MIME type sniffing
X-Frame-Options	`DENY`	Prevent clickjacking
Referrer-Policy	`strict-origin-when-cross-origin`	Limit referrer information leakage

Brute Force Protection

Combine rate limiting per IP with per-account lockout and exponential backoff.

python

import math

BASE_DELAY_SECONDS = 1
MAX_DELAY_SECONDS = 900  # 15 minutes


def get_lockout_delay(failed_attempts: int) -> int:
    """Calculate exponential backoff delay based on failed attempts."""
    if failed_attempts < 3:
        return 0
    delay = min(BASE_DELAY_SECONDS * (2 ** (failed_attempts - 3)), MAX_DELAY_SECONDS)
    return int(delay)


def check_brute_force(user) -> bool:
    """Return True if the user is currently locked out."""
    if user.failed_login_attempts < 3:
        return False

    delay = get_lockout_delay(user.failed_login_attempts)
    if user.last_failed_login:
        unlock_time = user.last_failed_login + timedelta(seconds=delay)
        if datetime.now(timezone.utc) < unlock_time:
            return True
    return False

Logging Failed Attempts

Log all failed authentication attempts for security monitoring and incident response.

python

import logging

security_logger = logging.getLogger("security")


def log_failed_login(email: str, ip: str, reason: str):
    security_logger.warning(
        "Failed login attempt",
        extra={
            "email": email,
            "ip": ip,
            "reason": reason,
            "timestamp": datetime.now(timezone.utc).isoformat(),
        },
    )

Constraints

✅ Do

•Rate limit all authentication endpoints with appropriate thresholds.
•Enforce HTTPS in production with automatic HTTP-to-HTTPS redirects.
•Use SameSite=Strict on all authentication cookies.
•Log all failed authentication attempts with IP address and timestamp.
•Set security headers (HSTS, X-Content-Type-Options, X-Frame-Options) on every response.
•Implement per-account lockout with exponential backoff alongside per-IP rate limiting.

❌ Don't

•Rely only on client-side validation for any security control.
•Allow HTTP connections in production environments.
•Expose token contents, internal error details, or stack traces in error responses.
•Skip logging for failed authentication attempts.
•Use a single rate limit strategy in isolation; combine IP-based and account-based protections.
•Hard-code rate limit values; make them configurable via environment variables.

Output Format

Rate limit exceeded (429):

json

{ "error": "Too many requests. Try again later." }

CSRF failure (403):

json

{ "error": "CSRF validation failed" }

Successful request with security headers:

code

HTTP/1.1 200 OK
Strict-Transport-Security: max-age=31536000; includeSubDomains
X-Content-Type-Options: nosniff
X-Frame-Options: DENY
Referrer-Policy: strict-origin-when-cross-origin

Dependencies

•Managing Sessions and Tokens -- cookie configuration and token handling that these protections wrap around.
•Managing Flask Middleware -- middleware patterns for applying security headers and rate limiting globally.