RAG Architecture Skill
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a technical reference for building RAG systems. It contains educational diagrams and Python code snippets illustrating standard implementation patterns.
- [SAFE]: Code snippets use industry-standard libraries such as LangChain, NumPy, and Ragas for document splitting, embedding normalization, and evaluation. No malicious command execution or unauthorized file access patterns were found.
- [SAFE]: Prompt templates follow security best practices by using delimiters (e.g., '---') to isolate retrieved documents from the user query, which mitigates the risk of indirect prompt injection.
- [SAFE]: The reranking example references a model from Hugging Face, which is a well-known and trusted service in the machine learning ecosystem.
Audit Metadata