Back to Prompts

Security · Advanced

The Red Team

Security Auditing

Turns the AI against its own creation. Essential for security-conscious vibecoders to find vulnerabilities before deployment.

Use in Build Lab

01

Adversarial Prompting

Run this AFTER you have generated your feature.

Prompt
Act as a malicious actor / Red Team Security Researcher.

Task: Try to find input vectors in this API/Code that could lead to:
1. Crash (DoS)
2. Unauthorized Access (Privilege Escalation)
3. Data Corruption (SQL Injection / XSS)
4. PII Leakage

Propose specific exploits and then provide the patched code to prevent them.