grill-me
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied plans and designs to generate interview questions, creating a surface for malicious instructions to be embedded in untrusted data. 1. Ingestion points: The skill takes a plan or design as initial user input (SKILL.md). 2. Boundary markers: No explicit delimiters or warnings are used to tell the agent to ignore instructions embedded in the design document. 3. Capability inventory: The skill instructs the agent to dispatch subagents to search the codebase and environment for facts (SKILL.md). 4. Sanitization: No sanitization or filtering of the user-provided plan is mentioned.
- [COMMAND_EXECUTION]: The skill instructs the agent to perform automated lookups in the local codebase and environment via subagents when answering questions (SKILL.md), which involves accessing and reading local files and environment data.
Audit Metadata