grill-me
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from user inputs and the project codebase, creating a surface for indirect prompt injection.
- Ingestion points: The agent is explicitly instructed to analyze a user's "plan or design" and to "explore the codebase" (SKILL.md).
- Boundary markers: There are no explicit delimiters (e.g., XML tags or triple quotes) or specific instructions to ignore malicious directives that might be embedded in the codebase or user plans.
- Capability inventory: The skill leverages the agent's ability to read and search files within the local filesystem to answer questions.
- Sanitization: No sanitization, validation, or filtering mechanisms are defined for the content retrieved from the codebase.
Audit Metadata