PromptInjection - Test LLM Applications for Prompt Injection Vulnerabilities
Test LLM applications for prompt injection vulnerabilities including jailbreak attempts and guardrail bypass techniques
Tags
Updated: 2026-05-11Capabilities
Typical Inputs
Typical Outputs
What this skill does
- test LLM applications
- extract system prompts
- manipulate contexts
- bypass guardrails
- discover injection points
- execute multi-stage attacks
Inputs
- target LLM application
- written authorization document
- testing scope boundaries
- target URL or endpoint
Outputs
- vulnerability report
- security findings
- test notifications
Requirements
- authorization environment
- professional security experience
- penetration testing permission
- ethical compliance