PromptInjection - LLM Prompt Injection Security Testing
Test LLM applications for prompt injection vulnerabilities and jailbreak attempts
Tags
Updated: 2026-05-11What this skill does
- test prompt injection
- detect jailbreak attempts
- extract system prompts
- bypass guardrails
- identify injection points
- perform reconnaissance
- test direct injection
- test indirect injection
- conduct multi-stage attacks
- enumerate API endpoints
- inspect JavaScript
- extract DOM data
- upload documents
- poison RAG systems
Inputs
- target LLM application
- written authorization
- target URL or endpoint
Outputs
- vulnerability findings
- security assessment report
- test documentation
Requirements
- written authorization required
- legal compliance
- browser automation support