RAG
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill contains text within
security.mdthat describes prompt injection as a security risk ('IGNORE ALL PREVIOUS INSTRUCTIONS'). This is documented as an example for the user to guard against, not as a malicious instruction to the agent. No actual prompt injection attempts were detected. - [NO_CODE]: The skill consists entirely of Markdown documentation and static configuration. While it contains Python code snippets, these are provided for educational purposes within Markdown blocks and are not executed by the agent.
- [EXTERNAL_DOWNLOADS]: No automated downloads or remote resource fetching were detected. The skill references well-known libraries (LangChain) and services (OpenAI, Pinecone) in a purely informational context.
- [CREDENTIALS_UNSAFE]: No hardcoded API keys or secrets were detected. The skill correctly instructs users to manage access control and handle PII (Personally Identifiable Information) responsibly.
Audit Metadata