PromptInjection - LLM Prompt Injection Security Testing
Test LLM applications for prompt injection vulnerabilities including jailbreaks and guardrail bypasses
Tags
Updated: 2026-05-11Capabilities
Typical Inputs
Typical Outputs
What this skill does
- test prompt injection
- detect jailbreaks
- extract system prompts
- bypass guardrails
- perform direct injection
- perform indirect injection
- execute multi-stage attacks
- conduct reconnaissance
- enumerate API endpoints
- identify injection points
- analyze DOM
- inspect JavaScript
- test RAG systems
Inputs
- LLM application
- authorization document
- target system URL
- API endpoints
- RAG system
- test documents
Outputs
- security assessment report
- vulnerability findings
- test results
- attack surface map
Requirements
- written authorization
- target system access
- browser automation
- voice notification server