Backend security

Security auditing for code vulnerabilities (OWASP Top 10, XSS, SQL injection) and dependency scanning (pnpm audit, Snyk). Use when handling user input, adding authentication, before deployments, or resolving CVEs.

install
source · Clone the upstream repo
git clone https://github.com/sgcarstrends/sgcarstrends
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/sgcarstrends/sgcarstrends "$T" && mkdir -p ~/.claude/skills && cp -r "$T/.claude/skills/security" ~/.claude/skills/sgcarstrends-backend-security && rm -rf "$T"
manifest: .claude/skills/security/SKILL.md
source content

Security Skill

Dependency Scanning

pnpm audit

# Run audit
pnpm audit

# Only high/critical
pnpm audit --audit-level=high

# Auto-fix
pnpm audit --fix

# JSON report
pnpm audit --json > audit.json

Fix Vulnerabilities

Direct dependencies: Update version in

pnpm-workspace.yaml
catalog

Transitive dependencies:

# Find dependency chain
pnpm why vulnerable-package

# Use overrides as last resort
# package.json
{
  "pnpm": {
    "overrides": {
      "vulnerable-package": "^3.1.0"
    }
  }
}

Snyk

snyk auth          # Authenticate
snyk test          # Test for vulnerabilities
snyk monitor       # Monitor for new vulnerabilities
snyk fix           # Auto-fix

OWASP Top 10 Checks

1. Broken Access Control

// ❌ No authorization
export async function deletePost(postId: string) {
  await db.delete(posts).where(eq(posts.id, postId));
}

// ✅ With authorization
export async function deletePost(postId: string, userId: string) {
  const post = await db.query.posts.findFirst({ where: eq(posts.id, postId) });
  if (post.authorId !== userId) throw new Error("Unauthorized");
  await db.delete(posts).where(eq(posts.id, postId));
}

2. Injection Prevention

// ❌ SQL Injection
const query = `SELECT * FROM users WHERE id = ${userId}`;

// ✅ Parameterized query (Drizzle ORM)
const user = await db.query.users.findFirst({ where: eq(users.id, userId) });

3. XSS Prevention

React escapes content by default. When rendering HTML:

  • Sanitize with
    sanitize-html
    library before rendering
  • Never render untrusted content directly

4. Rate Limiting

import { Ratelimit } from "@upstash/ratelimit";
import { redis } from "@sgcarstrends/utils";

const ratelimit = new Ratelimit({
  redis,
  limiter: Ratelimit.slidingWindow(5, "15 m"),
});

export async function login(email: string, password: string, ip: string) {
  const { success } = await ratelimit.limit(ip);
  if (!success) throw new Error("Too many login attempts");
  return verifyCredentials(email, password);
}

5. Password Security

import bcrypt from "bcrypt";

// ✅ Hash passwords
const hashedPassword = await bcrypt.hash(password, 10);

// ✅ Strong password validation
const passwordSchema = z.string()
  .min(12)
  .regex(/[A-Z]/, "Must contain uppercase")
  .regex(/[a-z]/, "Must contain lowercase")
  .regex(/[0-9]/, "Must contain number")
  .regex(/[^A-Za-z0-9]/, "Must contain special character");

6. SSRF Prevention

// ❌ SSRF vulnerability
export async function fetchUrl(url: string) {
  return await fetch(url);
}

// ✅ Whitelist approach
const ALLOWED_DOMAINS = ["api.example.com", "data.gov.sg"];

export async function fetchUrl(url: string) {
  const parsedUrl = new URL(url);
  if (!ALLOWED_DOMAINS.includes(parsedUrl.hostname)) {
    throw new Error("Domain not allowed");
  }
  return await fetch(url);
}

Input Validation

import { z } from "zod";

const userInputSchema = z.object({
  name: z.string().min(1).max(100),
  email: z.string().email(),
  age: z.number().int().min(0).max(150),
});

export async function createUser(data: unknown) {
  const validated = userInputSchema.parse(data);
  // Now safe to use
}

CORS Configuration

// ❌ Too permissive
app.use(cors({ origin: "*" }));

// ✅ Whitelist specific origins
app.use(cors({
  origin: [
    "https://sgcarstrends.com",
    "https://staging.sgcarstrends.com",
    process.env.NODE_ENV === "development" ? "http://localhost:3001" : "",
  ].filter(Boolean),
  credentials: true,
}));

Security Headers

// next.config.js
const securityHeaders = [
  { key: "Strict-Transport-Security", value: "max-age=63072000; includeSubDomains; preload" },
  { key: "X-Frame-Options", value: "SAMEORIGIN" },
  { key: "X-Content-Type-Options", value: "nosniff" },
  { key: "X-XSS-Protection", value: "1; mode=block" },
  { key: "Referrer-Policy", value: "origin-when-cross-origin" },
];

module.exports = {
  async headers() {
    return [{ source: "/:path*", headers: securityHeaders }];
  },
};

Environment Variables

// ❌ Hardcoded secret
const apiKey = "sk_live_EXAMPLE_NOT_REAL";

// ✅ From environment with validation
import { z } from "zod";

const envSchema = z.object({
  API_KEY: z.string().min(1),
  DATABASE_URL: z.string().url(),
});

const env = envSchema.parse(process.env);

CI Integration

# .github/workflows/security.yml
name: Security Audit

on:
  push:
    branches: [main]
  schedule:
    - cron: '0 0 * * 1'  # Weekly

jobs:
  audit:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      - uses: pnpm/action-setup@v2
      - run: pnpm install
      - run: pnpm audit --audit-level=high

Security Checklist

  • All user input validated (Zod schemas)
  • SQL injection prevented (using ORM)
  • XSS prevented (React escaping, sanitization)
  • Authentication implemented correctly
  • Authorization checks in place
  • Passwords hashed (bcrypt/argon2)
  • Rate limiting configured
  • Security headers set
  • CORS configured properly
  • HTTPS enforced
  • Dependencies audited (pnpm audit)
  • Secrets in environment variables
  • Error messages don't leak info

References