grill-with-docs
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process user-supplied design plans and project context, creating a surface for indirect prompt injection through data processed during documentation generation.\n
- Ingestion points: User-provided designs, plans, and the
CONTEXT.mdfile within the repository.\n - Boundary markers: The skill lacks explicit instructions for the agent to use delimiters or ignore embedded instructions when reading user-provided design documents.\n
- Capability inventory: The skill is capable of writing and rewriting files (specifically ADRs), modifying the
CONTEXT.mdfile, and invoking modeling and review tools.\n - Sanitization: No input validation or sanitization mechanisms are defined to ensure that the content being transcribed into documentation does not contain malicious instructions.
Audit Metadata