llm-risk-assess

Threat modeling for RAG pipelines involves mapping data flows and integrations to identify security risks. This Skill applies the OWASP Top 10 for LLMs to evaluate vector database checks, plugin integrations, and data-disclosure paths within your retrieval pipeline.

This Skill handles LLM threat modeling across chatbots, AI agents, and RAG-powered applications. It maps pipeline architectures and plugin integrations to identify security risks, providing repeatable artifacts like threat models and risk-based remediation guidance.

Red-team testing for LLM applications produces structured artifacts including threat models, automated test results, and remediation checklists. These outputs provide concrete evidence of security risks and deliver risk-based mitigations for chatbot and RAG pipeline vulnerabilities.

OWASP LLM risk assessment covers vector database security checks as part of its automated testing suite. It evaluates your RAG components to surface data-disclosure risks and identify vulnerabilities within the vector database integrations of your LLM pipeline.