Garak
What ChatGPT, Claude, Gemini & Grok actually say · July 2026
The verdict
Garak appears in 1 AI-ranked category — best position #3 for ai red teaming and llm security testing tool.
An incredibly comprehensive open-source CLI scanner with a large, active registry of specialized probing modules for rapid prompt injection and data leakage discovery.
Claude The broadest ready-made probe library for LLM vulnerability scanning (hallucination, data leakage, jailbreaks, toxicity), simple one-command scans, research-grade rigor and free
GPT Mature open-source LLM vulnerability scanner with a large probe ecosystem, reproducible testing, broad model support, and strong value for baseline security assessments
What would move Garak up
- GPT Expand beyond model-centric probes into stateful end-to-end agent, tool, and RAG exploitation
- Claude Modernize reporting and remediation guidance — findings are raw and researcher-oriented, needing translation before security teams can act
- Gemini Lacks robust native support for complex multi-turn stateful conversations and multi-modal inputs.
Top alternatives per the models: Promptfoo · PyRIT · Mindgard · Giskard
Rankings are computed from what the models answer, re-polled continuously · raw reasoning shown verbatim · methodology