langchain-architecture
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates for creating agents that process external documents via
TextLoaderandDirectoryLoader, and interact with tools likesend_email. This architecture creates an inherent attack surface where instructions embedded in processed data could attempt to influence agent actions. - Ingestion points:
TextLoader('documents.txt')andDirectoryLoader('./docs')in SKILL.md. - Boundary markers: The provided examples do not include explicit delimiters or instructions to ignore embedded commands within the ingested text.
- Capability inventory: The skill patterns include network access (OpenAI, SerpAPI), file reading, and potential tool-based actions like sending emails.
- Sanitization: No sanitization or input validation logic is present in the architectural templates provided.
Audit Metadata