Simple LLM service identification - translate IP:Port to Ollama, vLLM, LiteLLM, or 60+ other AI services in seconds
LLM Attack Testing Toolkit is a structured methodology and mindset framework for testing Large Language Model (LLM) applications against logic abuse, prompt injection, jailbreaks, and workflow manipulation.
Security scanner for local LLMs scanning LLM vulnerabilities including jailbreaks, prompt injection, training data leakage, and adversarial abuse
Autonomous offensive security framework for penetration testing. Golang harness plans and executes bounded tasks under an operator-owned ROE policy, with scope and action authority held in code rather than the model. Ships a hybrid-retrieval security knowledge base served by on-device MLX models.
Companion code for 'LLM Penetration Testing & AI Security' (2026): 105 tested Python examples — attacks, defenses, detection — plus an offline DVLLM Docker lab. Arc PI Taxonomy: 1,039,500 attack combinations.
A privacy-first LLM agent for automated vulnerability discovery, validation, and patching using local models.
A Claude Code plugin that drives an authorized AI/LLM security assessment end-to-end — black, gray, and white-box — over the industry red-team tools (garak, promptfoo, PyRIT, DeepTeam, IBM ART), then maps every finding to OWASP, MITRE ATLAS, and NIST.
To associate your repository with the llm-pentesting topic, visit your repo's landing page and select "manage topics."