---

name: security-review description: | Security vulnerability detection skill for reviewer agents. OWASP Top 10 checklist, secrets detection, input validation, authentication patterns. CRITICAL for code handling user input, authentication, APIs, or sensitive data.

Security Review

Security vulnerability detection and remediation guidance for ultrawork verification.

What This Skill Provides

Comprehensive security checklists covering:

• OWASP Top 10 vulnerabilities
• Secrets and credential exposure
• Input validation and injection prevention
• Authentication and authorization patterns
• API security best practices

When to Use This Skill

ALWAYS review when code involves:

• User input handling
• Authentication or authorization
• Database queries
• API endpoints
• File uploads
• Payment/financial operations
• External API integrations
• Sensitive data (PII, credentials, tokens)

IMMEDIATELY review when:

• Production incident occurred
• Dependency has known CVE
• User reports security concern
• Before major releases

OWASP Top 10 Checklist

1. Injection (SQL, NoSQL, Command)

Checklist:

Copy[ ] Queries use parameterization (no string concatenation)
[ ] ORMs used safely (no raw SQL strings)
[ ] User input is sanitized before queries
[ ] Command injection prevented (no shell execution with user input)
[ ] LDAP injection prevented
[ ] XPath injection prevented

Examples:

Copy// ❌ CRITICAL: SQL injection vulnerability
const query = `SELECT * FROM users WHERE id = ${userId}`;
await db.query(query);

// ✅ CORRECT: Parameterized query
const { data } = await supabase
  .from('users')
  .select('*')
  .eq('id', userId);

// ❌ CRITICAL: Command injection
const { exec } = require('child_process');
exec(`ping ${userInput}`, callback);

// ✅ CORRECT: Use libraries instead
const dns = require('dns');
dns.lookup(userInput, callback);

2. Broken Authentication

Checklist:

Copy[ ] Passwords are hashed (bcrypt, argon2, not MD5/SHA1)
[ ] JWT tokens properly validated
[ ] Sessions have timeouts
[ ] Multi-factor authentication available
[ ] Password reset tokens are single-use
[ ] Account lockout after failed attempts
[ ] Session fixation prevented
[ ] Credentials not in URLs or logs

Examples:

Copy// ❌ CRITICAL: Plaintext password comparison
if (password === storedPassword) { /* login */ }

// ✅ CORRECT: Hashed password comparison
import bcrypt from 'bcrypt';
const isValid = await bcrypt.compare(password, hashedPassword);

// ❌ CRITICAL: JWT not validated
const user = jwt.decode(token); // No verification!

// ✅ CORRECT: JWT validated
const user = jwt.verify(token, process.env.JWT_SECRET);

3. Sensitive Data Exposure

Checklist:

Copy[ ] HTTPS enforced (no HTTP endpoints)
[ ] Secrets in environment variables (not hardcoded)
[ ] PII encrypted at rest
[ ] Logs don't contain sensitive data
[ ] API keys not exposed in client code
[ ] Database credentials not committed
[ ] Encryption uses strong algorithms (AES-256, not DES)
[ ] Sensitive data not in URLs or query params

Examples:

Copy// ❌ CRITICAL: Hardcoded secret
const apiKey = "sk-proj-abc123xyz";

// ✅ CORRECT: Environment variable
const apiKey = process.env.OPENAI_API_KEY;
if (!apiKey) {
  throw new Error('OPENAI_API_KEY not configured');
}

// ❌ CRITICAL: Logging sensitive data
console.log('User login:', { email, password, apiKey });

// ✅ CORRECT: Sanitized logs
console.log('User login:', {
  email: email.replace(/(?<=.).(?=.*@)/g, '*'),
  passwordProvided: !!password
});

4. XML External Entities (XXE)

Checklist:

Copy[ ] XML parsers have external entity processing disabled
[ ] XML input is validated
[ ] Alternative formats used when possible (JSON)

5. Broken Access Control

Checklist:

Copy[ ] Authorization checked on every route
[ ] Users can only access their own resources
[ ] CORS configured properly (not '*' in production)
[ ] Object references are indirect (UUIDs, not sequential IDs)
[ ] Admin functions require admin role
[ ] Rate limiting prevents abuse

Examples:

Copy// ❌ CRITICAL: No authorization check
app.get('/api/user/:id', async (req, res) => {
  const user = await getUser(req.params.id);
  res.json(user); // Any user can see any profile!
});

// ✅ CORRECT: Verify access
app.get('/api/user/:id', authenticateUser, async (req, res) => {
  if (req.user.id !== req.params.id && !req.user.isAdmin) {
    return res.status(403).json({ error: 'Forbidden' });
  }
  const user = await getUser(req.params.id);
  res.json(user);
});

6. Security Misconfiguration

Checklist:

Copy[ ] Default credentials changed
[ ] Debug mode disabled in production
[ ] Error messages don't expose internals
[ ] Security headers set (CSP, HSTS, X-Frame-Options)
[ ] Unnecessary features/ports disabled
[ ] Software is up to date
[ ] Directory listing disabled
[ ] Detailed errors not shown to users

Security Headers:

Copy// Required security headers
app.use((req, res, next) => {
  res.setHeader('X-Content-Type-Options', 'nosniff');
  res.setHeader('X-Frame-Options', 'DENY');
  res.setHeader('Content-Security-Policy', "default-src 'self'");
  res.setHeader('Strict-Transport-Security', 'max-age=31536000');
  next();
});

7. Cross-Site Scripting (XSS)

Checklist:

Copy[ ] Output is escaped/sanitized
[ ] Content-Security-Policy header set
[ ] Frameworks escape by default (React, Vue)
[ ] innerHTML avoided (use textContent)
[ ] User input not directly rendered
[ ] URL parameters sanitized

Examples:

Copy// ❌ HIGH: XSS vulnerability
element.innerHTML = userInput;

// ✅ CORRECT: Use textContent
element.textContent = userInput;

// ✅ CORRECT: Sanitize if HTML needed
import DOMPurify from 'dompurify';
element.innerHTML = DOMPurify.sanitize(userInput);

8. Insecure Deserialization

Checklist:

Copy[ ] User input not deserialized without validation
[ ] JSON.parse has try/catch
[ ] Deserialization libraries up to date
[ ] Type checking after deserialization

9. Using Components with Known Vulnerabilities

Checklist:

Copy[ ] Dependencies up to date
[ ] npm audit clean (no high/critical issues)
[ ] CVEs monitored
[ ] Automated dependency scanning enabled
[ ] Vulnerable packages replaced

Commands:

Copy# Check for vulnerable dependencies
npm audit

# Fix automatically
npm audit fix

# High severity only
npm audit --audit-level=high

10. Insufficient Logging & Monitoring

Checklist:

Copy[ ] Security events logged (login, failed auth, access control)
[ ] Logs monitored
[ ] Alerts configured for anomalies
[ ] Logs immutable (append-only)
[ ] Log retention policy defined
[ ] Sensitive data not logged

Additional Security Patterns

Server-Side Request Forgery (SSRF)

Checklist:

Copy[ ] URLs validated and whitelisted
[ ] Internal IPs blocked
[ ] URL scheme restricted (http/https only)

Example:

Copy// ❌ HIGH: SSRF vulnerability
const response = await fetch(userProvidedUrl);

// ✅ CORRECT: Validate and whitelist
const allowedDomains = ['api.example.com', 'cdn.example.com'];
const url = new URL(userProvidedUrl);
if (!allowedDomains.includes(url.hostname)) {
  throw new Error('Invalid URL');
}
const response = await fetch(url.toString());

Race Conditions (Financial/Critical Operations)

Checklist:

Copy[ ] Atomic transactions for financial operations
[ ] Row-level locking for concurrent updates
[ ] Balance checks inside transactions
[ ] Idempotency keys for retries

Example:

Copy// ❌ CRITICAL: Race condition in balance check
const balance = await getBalance(userId);
if (balance >= amount) {
  await withdraw(userId, amount); // Another request could withdraw in parallel!
}

// ✅ CORRECT: Atomic transaction with lock
await db.transaction(async (trx) => {
  const balance = await trx('balances')
    .where({ user_id: userId })
    .forUpdate() // Lock row
    .first();

  if (balance.amount < amount) {
    throw new Error('Insufficient balance');
  }

  await trx('balances')
    .where({ user_id: userId })
    .decrement('amount', amount);
});

Rate Limiting

Checklist:

Copy[ ] Rate limiting on authentication endpoints
[ ] Rate limiting on API endpoints
[ ] Rate limiting on expensive operations
[ ] Different limits for authenticated/anonymous users

Example:

Copyimport rateLimit from 'express-rate-limit';

const authLimiter = rateLimit({
  windowMs: 15 * 60 * 1000, // 15 minutes
  max: 5, // 5 attempts
  message: 'Too many login attempts, please try again later'
});

app.post('/api/login', authLimiter, loginHandler);

Secrets Detection

Patterns to Detect

Copy# Search for potential secrets
grep -r "api[_-]?key\|password\|secret\|token" \
  --include="*.js" \
  --include="*.ts" \
  --include="*.json" .

# Check git history for committed secrets
git log -p | grep -i "password\|api_key\|secret"

Common secret patterns:

• API keys: sk-, pk-, api_key=
• Passwords: password =, pwd=
• Tokens: token=, access_token=
• Private keys: -----BEGIN PRIVATE KEY-----

Allowed patterns (not secrets):

• .env.example files
• Test credentials clearly marked
• Public API keys (Stripe publishable key)
• Hash outputs (not keys)

Security Review Process

Step 1: Automated Scanning

Copy# Dependency audit
npm audit

# Security linting
npx eslint . --plugin security

# Secrets scanning
npx trufflehog filesystem . --json

Step 2: Manual Code Review

Focus on:

• Authentication/authorization code
• User input handling
• Database queries
• API endpoints
• File operations

Step 3: Threat Modeling

Ask:

• What could an attacker do with this endpoint?
• What happens if user provides malicious input?
• Can users access others' data?
• Are financial operations atomic?

Step 4: Evidence Collection

Copy# Add security assessment to evidence
bun "{SCRIPTS_PATH}/task-update.js" --session ${CLAUDE_SESSION_ID} --id verify \
  --add-evidence "Security Review: PASS" \
  --add-evidence "- No hardcoded secrets" \
  --add-evidence "- Input validation present" \
  --add-evidence "- Authorization checked" \
  --add-evidence "- npm audit clean"

Severity Classification

Common False Positives

Not every finding is a vulnerability:

• Environment variables in .env.example (examples, not secrets)
• Test credentials in test files (if clearly marked)
• Public API keys (Stripe publishable key)
• SHA256/MD5 for checksums (not passwords)
• Constants that look like secrets but aren't

Always verify context before flagging.

Emergency Response

If CRITICAL vulnerability found:

1. Document - Create detailed report
2. Notify - Alert project owner immediately
3. Recommend Fix - Provide secure code example
4. Test Fix - Verify remediation works
5. Verify Impact - Check if vulnerability was exploited
6. Rotate Secrets - If credentials exposed
7. Update Docs - Add to security knowledge base

Integration with Ultrawork

During VERIFICATION phase:

Copy# Run security review
reviewer --task-id 1 --focus security

# Block if critical issues found
bun "{SCRIPTS_PATH}/task-update.js" --session ${CLAUDE_SESSION_ID} --id verify \
  --status resolved \
  --add-evidence "VERDICT: FAIL - CRITICAL security issue found" \
  --add-evidence "Hardcoded API key in src/api.ts:42"

Quick Reference

Must-check security items:

• No hardcoded secrets
• Input validation on all user data
• Parameterized queries (no SQL injection)
• Authorization on protected routes
• HTTPS enforced
• Dependencies up to date

Critical patterns:

• SQL injection
• XSS vulnerabilities
• Authentication bypass
• Authorization bypass
• Hardcoded credentials
• Command injection