# Prompt Injection via Untrusted Input

- ID: javascript-prompt-injection
- Severity: HIGH
- CWE: Injection (CWE-74)
- Languages: JavaScript, TypeScript
- Frameworks: express, fastify

## Description

Detects user input flowing directly into AI/LLM prompts without sanitization.

## Detection Message

Untrusted input from {source} flows directly into AI prompt at {sink}.
This allows an attacker to manipulate AI responses (prompt injection attack).

## Remediation

Use system prompts and sanitize user input with length limits before including in prompts.

```javascript
const sanitized = userInput.substring(0, 500);
const messages = [
  { role: 'system', content: 'Answer only about products.' },
  { role: 'user', content: sanitized }
];
```

Learn more: https://shoulder.dev/learn/javascript/cwe-74/prompt-injection

## Documentation

[object Object]

## Related Rules

- **AI Prompt Injection** [HIGH]: 
- **AI Prompt Injection** [HIGH]: