production-rag-checklist
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHNO_CODE
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill provides code templates that ingest untrusted user queries via a FastAPI endpoint ('request.query'). The snippets lack boundary markers and sanitization and involve LLM invocation ('rag_pipeline.ainvoke'); however, the skill is a static documentation resource (NO_CODE) and does not execute these templates or process external data itself.
- [Unverifiable Dependencies & Remote Code Execution] (INFO): References standard Python libraries (fastapi, pydantic, redis, structlog) are present in documentation blocks. No active downloads or runtime execution of these packages occurs.
- [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials or sensitive file access patterns were detected. The 'logger.info' alert from the scanner is a false positive on a standard Python method call.
- [Metadata Poisoning] (SAFE): Metadata is consistent with the skill's purpose as a production deployment checklist.
Recommendations
- Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata