developer coding user risk: medium

Secure User Login Code Generator

Instructs to write code featuring user login with passwords stored salted and protected strongly in the database, and robust security measures for both backend and frontend.

Policy sensitive
Human review

PROMPT

Yazılacak kod aşağıdaki yeteneklerde olacak.

1. kullanıcı girişi olacak ve kullanıcı şifresi veritabanında salt ve diğer güçlü şifre korumaları ile tutulacak.
2. backend ve frontend güçlü güvenlik sıkılaştırmalarına sahip olacak.

EXPECTED OUTPUT

Format: code

SUCCESS CRITERIA

Implement user login functionality with passwords stored in database using salt and other strong protections.
Ensure backend and frontend have strong security measures.

FAILURE MODES

Vague specifications may lead to incomplete or insecure implementations.
No programming language or framework specified.
No details on exact security practices beyond high-level mentions.

CAVEATS

Missing context

Programming language and framework for backend/frontend.
Database type.
Specific security measures (e.g., hashing algorithm, CSRF protection, input validation).
Output format (e.g., full application code, snippets, or architecture).
Scope (e.g., full app or just authentication module).

Ambiguities

Unclear what 'other strong password protections' means beyond salting.
Vague on what constitutes 'strong security tightenings' for backend and frontend.

QUALITY

OVERALL: 0.45
CLARITY: 0.75
SPECIFICITY: 0.40
REUSABILITY: 0.30
COMPLETENESS: 0.35

IMPROVEMENT SUGGESTIONS

Specify tech stack: 'Use Node.js/Express for backend, React for frontend, PostgreSQL for database.'
Detail security: 'Passwords hashed with bcrypt (including salt), implement JWT auth, helmet.js, CORS, input sanitization, HTTPS.'
Add output requirements: 'Provide complete, runnable code with setup instructions and security explanations.'
Define success criteria: 'Code must pass basic security scans like npm audit.'

USAGE

Copy the prompt above and paste it into your AI of choice — Claude, ChatGPT, Gemini, or anywhere else you're working. Replace any placeholder sections with your own context, then ask for the output.