Security hardening

Security Hardening Patterns

HTTP Security Headers

const securityHeaders = {
  'Content-Security-Policy': "default-src 'self'; script-src 'self'; style-src 'self' 'unsafe-inline'; img-src 'self' data: https:; font-src 'self'; connect-src 'self'; frame-ancestors 'none'",
  'X-Content-Type-Options': 'nosniff',
  'X-Frame-Options': 'DENY',
  'X-XSS-Protection': '0', // Disabled — CSP is the modern replacement
  'Referrer-Policy': 'strict-origin-when-cross-origin',
  'Permissions-Policy': 'camera=(), microphone=(), geolocation=()',
  'Strict-Transport-Security': 'max-age=63072000; includeSubDomains; preload',
};

CORS

const corsOptions = {
  origin: ['https://app.example.com'], // Never use '*' with credentials
  methods: ['GET', 'POST', 'PUT', 'DELETE'],
  credentials: true,
  maxAge: 86400,
};

• Never use origin: '*' with credentials: true
• Whitelist specific origins, never reflect the Origin header
• Set maxAge to reduce preflight requests

Authentication (Session-Based)

// Session cookie settings
const sessionCookie = {
  httpOnly: true,    // No JavaScript access
  secure: true,      // HTTPS only
  sameSite: 'lax',   // CSRF protection
  maxAge: 86400,     // 24 hours
  path: '/',
};

// Password hashing (bcrypt, 12+ rounds)
import { hash, compare } from 'bcrypt';
const ROUNDS = 12;
const hashed = await hash(password, ROUNDS);
const valid = await compare(input, hashed);

• Session-only auth (no JWT in cookies — JWTs can't be revoked)
• bcrypt with 12+ rounds (not MD5, SHA-256, or plain text)
• httpOnly cookies prevent XSS token theft
• sameSite=lax prevents CSRF on state-changing requests

Input Validation

import { z } from 'zod';

const CreateUserSchema = z.object({
  email: z.string().email().max(255),
  name: z.string().min(1).max(100),
  role: z.enum(['admin', 'user']),
});

// Validate at system boundaries
export async function createUser(input: unknown) {
  const data = CreateUserSchema.parse(input); // throws on invalid
  // ... safe to use data
}

• Validate ALL external input (user input, API requests, webhooks)
• Use Zod schemas — never trust typeof or manual checks
• Validate at system boundaries, trust internal code
• Set reasonable max lengths to prevent abuse

Rate Limiting

// Per-endpoint rate limiting
const rateLimiter = {
  '/api/auth/sign-in': { windowMs: 15 * 60 * 1000, max: 10 },  // 10 per 15 min
  '/api/auth/sign-up': { windowMs: 60 * 60 * 1000, max: 5 },   // 5 per hour
  '/api/webhooks':     { windowMs: 60 * 1000, max: 100 },       // 100 per minute
  '/api/*':            { windowMs: 60 * 1000, max: 300 },        // 300 per minute (default)
};

XSS Prevention

• Never use dangerouslySetInnerHTML without sanitization
• Validate URLs: block javascript:, vbscript:, data: schemes
• Use CSP to prevent inline script execution
• Escape user content in HTML attributes

function isSafeUrl(url: string): boolean {
  try {
    const parsed = new URL(url);
    return ['http:', 'https:', 'mailto:'].includes(parsed.protocol);
  } catch {
    return false;
  }
}

SQL Injection Prevention

• Use parameterized queries (Drizzle ORM does this automatically)
• Never concatenate user input into SQL strings
• Use sql.placeholder() for dynamic queries

// Safe (parameterized)
const users = await db.select().from(usersTable).where(eq(usersTable.email, userInput));

// DANGEROUS — never do this
const users = await db.execute(`SELECT * FROM users WHERE email = '${userInput}'`);

Secrets Management

• Never commit .env files, API keys, or credentials
• Prefer a single encrypted vault (or platform secret store) as source of truth; inject at runtime
• RevealUI Studio / RevFleet: secrets live in revvault (revvault run --env KEY=path -- cmd / with-secrets). Env vars are a downstream injection surface, not the store. See fleet secrets hardline. Lookup and inject: revvault-resolve.
• Generic apps: environment variables or managed secret stores; never hardcode
• Rotate credentials after any suspected exposure
• Scan for secrets in CI (Gitleaks, GitHub Secret Scanning)

OWASP Top 10 Checklist

1. Broken Access Control — enforce on every endpoint, not just UI
2. Cryptographic Failures — bcrypt for passwords, TLS everywhere
3. Injection — parameterized queries, input validation
4. Insecure Design — threat model before coding
5. Security Misconfiguration — security headers, disable debug in prod
6. Vulnerable Components — audit dependencies, pin versions
7. Authentication Failures — rate limit, brute force protection
8. Data Integrity Failures — verify webhook signatures, validate JWTs
9. Logging Failures — log auth events, don't log secrets
10. SSRF — validate URLs, block internal networks

---

Skill by RevealUI Studio — the agentic business runtime.