Rodin
647928a0a1
Fundamentals: secure-defaults, input-validation, credential-handling, audit-logging Identity: authentication, authorization Attack Prevention: injection-prevention, dos-prevention, prompt-injection
Security Patterns
Scannable patterns for security code review. Each file has:
- Rule — what to do
- Correct Pattern — code that works (Python)
- Incorrect Pattern — common mistakes
- Edge Cases — gotchas
Patterns
Fundamentals
| File | Topic |
|---|---|
| secure-defaults.md | Fail closed, deny by default, defense in depth |
| input-validation.md | Allowlist > blocklist, validate at boundaries |
| credential-handling.md | No hardcoded secrets, environment/secret manager |
| audit-logging.md | What to log, what not to log |
Identity
| File | Topic |
|---|---|
| authentication.md | Passwords, tokens, MFA, brute force protection |
| authorization.md | Permission checks, IDOR prevention, privilege escalation |
Attack Prevention
| File | Topic |
|---|---|
| injection-prevention.md | SQL, command, template, path traversal |
| dos-prevention.md | Rate limiting, resource bounds, algorithmic complexity |
| prompt-injection.md | LLM security, data/instruction separation |
Sources
Usage
Reference these patterns when building or reviewing systems. Code examples are in Python for universal model comprehension; concepts apply to any language.