eval-mcp
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided queries to search an external source (the handbook) and incorporates the results into the final response, creating a surface for potential instruction injection from the retrieved content.
- Ingestion Point: The user's lookup request is passed as a
queryargument to themcp__eval-mcp__handbook-searchtool as described inSKILL.md. - Boundary Markers: No explicit delimiters or instructions are provided to the agent to treat the retrieved handbook content as data rather than instructions.
- Capability Inventory: The skill facilitates the execution of the
mcp__eval-mcp__handbook-searchtool. - Sanitization: The skill does not define any sanitization or validation logic for the data returned from the search tool before it is used to generate the final answer.
Audit Metadata