rag-patterns
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a RAG architecture that processes untrusted external documents and user queries, establishing an attack surface for indirect prompt injection. \n
- Ingestion points: The skill parses external files and user input as detailed in
SKILL.mdandreferences/multimodal-memory-and-caching.md. \n - Boundary markers: The provided templates do not specify the use of delimiters or instructions for the agent to ignore instructions embedded in retrieved content. \n
- Capability inventory: The skill involves file extraction, vector search, and LLM interaction. \n
- Sanitization: No explicit sanitization or validation logic is present in the provided Python code snippets.
Audit Metadata