security-review

Name: security-review
Author: affaan-m

byaffaan-m· 86 skills

Installs

12k

Stars

234k

Forks

36k

How to install security-review?

affaan-m/everything-claude-code/security-review

$npx -y skills add affaan-m/everything-claude-code --skill security-review

Installs into the current project.

›Prefer a prompt? Paste this to your agent

Use this skill

Run `npx skills use "https://github.com/affaan-m/everything-claude-code" --skill "affaan-m/everything-claude-code/security-review"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the whole pack

Use the skills in "https://github.com/affaan-m/everything-claude-code" that are relevant to the current task. Run `npx skills add "https://github.com/affaan-m/everything-claude-code"` and select the relevant skills, then follow their instructions.

Files · 1

View on GitHub

SKILL.md

1	# Security Review Skill
2
3	This skill ensures all code follows security best practices and identifies potential vulnerabilities.
4
5	## When to Activate
6
7	- Implementing authentication or authorization
8	- Handling user input or file uploads
9	- Creating new API endpoints
10	- Working with secrets or credentials
11	- Implementing payment features
12	- Storing or transmitting sensitive data
13	- Integrating third-party APIs
14
15	## Security Checklist
16
17	### 1. Secrets Management
18
19	#### FAIL: NEVER Do This
20	```typescript
21	const apiKey = "sk-proj-xxxxx" // Hardcoded secret
22	const dbPassword = "password123" // In source code
23	```
24
25	#### PASS: ALWAYS Do This
26	```typescript
27	const apiKey = process.env.OPENAI_API_KEY
28	const dbUrl = process.env.DATABASE_URL
29
30	// Verify secrets exist
31	if (!apiKey) {
32	throw new Error('OPENAI_API_KEY not configured')
33	}
34	```
35
36	#### Verification Steps
37	- [ ] No hardcoded API keys, tokens, or passwords
38	- [ ] All secrets in environment variables
39	- [ ] `.env.local` in .gitignore
40	- [ ] No secrets in git history
41	- [ ] Production secrets in hosting platform (Vercel, Railway)
42
43	### 2. Input Validation
44
45	#### Always Validate User Input
46	```typescript
47	import { z } from 'zod'
48
49	// Define validation schema
50	const CreateUserSchema = z.object({
51	email: z.string().email(),
52	name: z.string().min(1).max(100),
53	age: z.number().int().min(0).max(150)
54	})
55
56	// Validate before processing
57	export async function createUser(input: unknown) {
58	try {
59	const validated = CreateUserSchema.parse(input)
60	return await db.users.create(validated)
61	} catch (error) {
62	if (error instanceof z.ZodError) {
63	return { success: false, errors: error.errors }
64	}
65	throw error
66	}
67	}
68	```
69
70	#### File Upload Validation
71	```typescript
72	function validateFileUpload(file: File) {
73	// Size check (5MB max)
74	const maxSize = 5 * 1024 * 1024
75	if (file.size > maxSize) {
76	throw new Error('File too large (max 5MB)')
77	}
78
79	// Type check
80	const allowedTypes = ['image/jpeg', 'image/png', 'image/gif']
81	if (!allowedTypes.includes(file.type)) {
82	throw new Error('Invalid file type')
83	}
84
85	// Extension check
86	const allowedExtensions = ['.jpg', '.jpeg', '.png', '.gif']
87	const extension = file.name.toLowerCase().match(/\.[^.]+$/)?.[0]
88	if (!extension \|\| !allowedExtensions.includes(extension)) {
89	throw new Error('Invalid file extension')
90	}
91
92	return true
93	}
94	```
95
96	#### Verification Steps
97	- [ ] All user inputs validated with schemas
98	- [ ] File uploads restricted (size, type, extension)
99	- [ ] No direct use of user input in queries
100	- [ ] Whitelist validation (not blacklist)
101	- [ ] Error messages don't leak sensitive info
102
103	### 3. SQL Injection Prevention
104
105	#### FAIL: NEVER Concatenate SQL
106	```typescript
107	// DANGEROUS - SQL Injection vulnerability
108	const query = `SELECT * FROM users WHERE email = '${userEmail}'`
109	await db.query(query)
110	```
111
112	#### PASS: ALWAYS Use Parameterized Queries
113	```typescript
114	// Safe - parameterized query
115	const { data } = await supabase
116	.from('users')
117	.select('*')
118	.eq('email', userEmail)
119
120	// Or with raw SQL
121	await db.query(
122	'SELECT * FROM users WHERE email = $1',
123	[userEmail]
124	)
125	```
126
127	#### Verification Steps
128	- [ ] All database queries use parameterized queries
129	- [ ] No string concatenation in SQL
130	- [ ] ORM/query builder used correctly
131	- [ ] Supabase queries properly sanitized
132
133	### 4. Authentication & Authorization
134
135	#### JWT Token Handling
136	```typescript
137	// FAIL: WRONG: localStorage (vulnerable to XSS)
138	localStorage.setItem('token', token)
139
140	// PASS: CORRECT: httpOnly cookies
141	res.setHeader('Set-Cookie',
142	`token=${token}; HttpOnly; Secure; SameSite=Strict; Max-Age=3600`)
143	```
144
145	#### Authorization Checks
146	```typescript
147	export async function deleteUser(userId: string, requesterId: string) {
148	// ALWAYS verify authorization first
149	const requester = await db.users.findUnique({
150	where: { id: requesterId }
151	})
152
153	if (requester.role !== 'admin') {
154	return NextResponse.json(
155	{ error: 'Unauthorized' },
156	{ status: 403 }
157	)
158	}
159
160	// Proceed with deletion
161	await db.users.delete({ where: { id: userId } })
162	}
163	```
164
165	#### Row Level Security (Supabase)
166	```sql
167	-- Enable RLS on all tables
168	ALTER TABLE users ENABLE ROW LEVEL SECURITY;
169
170	-- Users can only view their own data
171	CREATE POLICY "Users view own data"
172	ON users FOR SELECT
173	USING (auth.uid() = id);
174
175	-- Users can only update their own data
176	CREATE POLICY "Users update own data"
177	ON users FOR UPDATE
178	USING (auth.uid() = id);
179	```
180
181	#### Verification Steps
182	- [ ] Tokens stored in httpOnly cookies (not localStorage)
183	- [ ] Authorization checks before sensitive operations
184	- [ ] Row Level Security enabled in Supabase
185	- [ ] Role-based access control implemented
186	- [ ] Session management secure
187
188	### 5. XSS Prevention
189
190	#### Sanitize HTML
191	```typescript
192	import DOMPurify from 'isomorphic-dompurify'
193
194	// ALWAYS sanitize user-provided HTML
195	function renderUserContent(html: string) {
196	const clean = DOMPurify.sanitize(html, {
197	ALLOWED_TAGS: ['b', 'i', 'em', 'strong', 'p'],
198	ALLOWED_ATTR: []
199	})
200	return <div dangerouslySetInnerHTML={{ __html: clean }} />
201	}
202	```
203
204	#### Content Security Policy
205	```typescript
206	// next.config.js
207	const securityHeaders = [
208	{
209	key: 'Content-Security-Policy',
210	value: `
211	default-src 'self';
212	script-src 'self' 'unsafe-eval' 'unsafe-inline';
213	style-src 'self' 'unsafe-inline';
214	img-src 'self' data: https:;
215	font-src 'self';
216	connect-src 'self' https://api.example.com;
217	`.replace(/\s{2,}/g, ' ').trim()
218	}
219	]
220	```
221
222	#### Verification Steps
223	- [ ] User-provided HTML sanitized
224	- [ ] CSP headers configured
225	- [ ] No unvalidated dynamic content rendering
226	- [ ] React's built-in XSS protection used
227
228	### 6. CSRF Protection
229
230	#### CSRF Tokens
231	```typescript
232	import { csrf } from '@/lib/csrf'
233
234	export async function POST(request: Request) {
235	const token = request.headers.get('X-CSRF-Token')
236
237	if (!csrf.verify(token)) {
238	return NextResponse.json(
239	{ error: 'Invalid CSRF token' },
240	{ status: 403 }
241	)
242	}
243
244	// Process request
245	}
246	```
247
248	#### SameSite Cookies
249	```typescript
250	res.setHeader('Set-Cookie',
251	`session=${sessionId}; HttpOnly; Secure; SameSite=Strict`)
252	```
253
254	#### Verification Steps
255	- [ ] CSRF tokens on state-changing operations
256	- [ ] SameSite=Strict on all cookies
257	- [ ] Double-submit cookie pattern implemented
258
259	### 7. Rate Limiting
260
261	#### API Rate Limiting
262	```typescript
263	import rateLimit from 'express-rate-limit'
264
265	const limiter = rateLimit({
266	windowMs: 15 * 60 * 1000, // 15 minutes
267	max: 100, // 100 requests per window
268	message: 'Too many requests'
269	})
270
271	// Apply to routes
272	app.use('/api/', limiter)
273	```
274
275	#### Expensive Operations
276	```typescript
277	// Aggressive rate limiting for searches
278	const searchLimiter = rateLimit({
279	windowMs: 60 * 1000, // 1 minute
280	max: 10, // 10 requests per minute
281	message: 'Too many search requests'
282	})
283
284	app.use('/api/search', searchLimiter)
285	```
286
287	#### Verification Steps
288	- [ ] Rate limiting on all API endpoints
289	- [ ] Stricter limits on expensive operations
290	- [ ] IP-based rate limiting
291	- [ ] User-based rate limiting (authenticated)
292
293	### 8. Sensitive Data Exposure
294
295	#### Logging
296	```typescript
297	// FAIL: WRONG: Logging sensitive data
298	console.log('User login:', { email, password })
299	console.log('Payment:', { cardNumber, cvv })
300
301	// PASS: CORRECT: Redact sensitive data
302	console.log('User login:', { email, userId })
303	console.log('Payment:', { last4: card.last4, userId })
304	```
305
306	#### Error Messages
307	```typescript
308	// FAIL: WRONG: Exposing internal details
309	catch (error) {
310	return NextResponse.json(
311	{ error: error.message, stack: error.stack },
312	{ status: 500 }
313	)
314	}
315
316	// PASS: CORRECT: Generic error messages
317	catch (error) {
318	console.error('Internal error:', error)
319	return NextResponse.json(
320	{ error: 'An error occurred. Please try again.' },
321	{ status: 500 }
322	)
323	}
324	```
325
326	#### Verification Steps
327	- [ ] No passwords, tokens, or secrets in logs
328	- [ ] Error messages generic for users
329	- [ ] Detailed errors only in server logs
330	- [ ] No stack traces exposed to users
331
332	### 9. Blockchain Security (Solana)
333
334	#### Wallet Verification
335	```typescript
336	import { verify } from '@solana/web3.js'
337
338	async function verifyWalletOwnership(
339	publicKey: string,
340	signature: string,
341	message: string
342	) {
343	try {
344	const isValid = verify(
345	Buffer.from(message),
346	Buffer.from(signature, 'base64'),
347	Buffer.from(publicKey, 'base64')
348	)
349	return isValid
350	} catch (error) {
351	return false
352	}
353	}
354	```
355
356	#### Transaction Verification
357	```typescript
358	async function verifyTransaction(transaction: Transaction) {
359	// Verify recipient
360	if (transaction.to !== expectedRecipient) {
361	throw new Error('Invalid recipient')
362	}
363
364	// Verify amount
365	if (transaction.amount > maxAmount) {
366	throw new Error('Amount exceeds limit')
367	}
368
369	// Verify user has sufficient balance
370	const balance = await getBalance(transaction.from)
371	if (balance < transaction.amount) {
372	throw new Error('Insufficient balance')
373	}
374
375	return true
376	}
377	```
378
379	#### Verification Steps
380	- [ ] Wallet signatures verified
381	- [ ] Transaction details validated
382	- [ ] Balance checks before transactions
383	- [ ] No blind transaction signing
384
385	### 10. Dependency Security
386
387	#### Regular Updates
388	```bash
389	# Check for vulnerabilities
390	npm audit
391
392	# Fix automatically fixable issues
393	npm audit fix
394
395	# Update dependencies
396	npm update
397
398	# Check for outdated packages
399	npm outdated
400	```
401
402	#### Lock Files
403	```bash
404	# ALWAYS commit lock files
405	git add package-lock.json
406
407	# Use in CI/CD for reproducible builds
408	npm ci # Instead of npm install
409	```
410
411	#### Verification Steps
412	- [ ] Dependencies up to date
413	- [ ] No known vulnerabilities (npm audit clean)
414	- [ ] Lock files committed
415	- [ ] Dependabot enabled on GitHub
416	- [ ] Regular security updates
417
418	## Security Testing
419
420	### Automated Security Tests
421	```typescript
422	// Test authentication
423	test('requires authentication', async () => {
424	const response = await fetch('/api/protected')
425	expect(response.status).toBe(401)
426	})
427
428	// Test authorization
429	test('requires admin role', async () => {
430	const response = await fetch('/api/admin', {
431	headers: { Authorization: `Bearer ${userToken}` }
432	})
433	expect(response.status).toBe(403)
434	})
435
436	// Test input validation
437	test('rejects invalid input', async () => {
438	const response = await fetch('/api/users', {
439	method: 'POST',
440	body: JSON.stringify({ email: 'not-an-email' })
441	})
442	expect(response.status).toBe(400)
443	})
444
445	// Test rate limiting
446	test('enforces rate limits', async () => {
447	const requests = Array(101).fill(null).map(() =>
448	fetch('/api/endpoint')
449	)
450
451	const responses = await Promise.all(requests)
452	const tooManyRequests = responses.filter(r => r.status === 429)
453
454	expect(tooManyRequests.length).toBeGreaterThan(0)
455	})
456	```
457
458	## Pre-Deployment Security Checklist
459
460	Before ANY production deployment:
461
462	- [ ] Secrets: No hardcoded secrets, all in env vars
463	- [ ] Input Validation: All user inputs validated
464	- [ ] SQL Injection: All queries parameterized
465	- [ ] XSS: User content sanitized
466	- [ ] CSRF: Protection enabled
467	- [ ] Authentication: Proper token handling
468	- [ ] Authorization: Role checks in place
469	- [ ] Rate Limiting: Enabled on all endpoints
470	- [ ] HTTPS: Enforced in production
471	- [ ] Security Headers: CSP, X-Frame-Options configured
472	- [ ] Error Handling: No sensitive data in errors
473	- [ ] Logging: No sensitive data logged
474	- [ ] Dependencies: Up to date, no vulnerabilities
475	- [ ] Row Level Security: Enabled in Supabase
476	- [ ] CORS: Properly configured
477	- [ ] File Uploads: Validated (size, type)
478	- [ ] Wallet Signatures: Verified (if blockchain)
479
480	## Resources
481
482	- [OWASP Top 10](https://owasp.org/www-project-top-ten/)
483	- [Next.js Security](https://nextjs.org/docs/security)
484	- [Supabase Security](https://supabase.com/docs/guides/auth)
485	- [Web Security Academy](https://portswigger.net/web-security)
486
487	---
488
489	Remember: Security is not optional. One vulnerability can compromise the entire platform. When in doubt, err on the side of caution.

6 picks

Type

Run `npx skills use "https://github.com/affaan-m/everything-claude-code" --skill "affaan-m/everything-claude-code/security-review"` and follow the generated skill instructions now. Read its complete output, redirecting it to a temporary file first if necessary. Resolve relative paths from the supporting-files directory it provides.

Use the skills in "https://github.com/affaan-m/everything-claude-code" that are relevant to the current task. Run `npx skills add "https://github.com/affaan-m/everything-claude-code"` and select the relevant skills, then follow their instructions.

security-review

How to install security-review?

Related

security-review

How to install security-review?

Related