interview-me
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted user input from interview rounds and use it to dynamically adjust the agent's plan.
- Ingestion points: User responses provided through the
AskUserQuestioninterface as described inSKILL.md. - Boundary markers: None; the instructions do not specify any delimiters or safety warnings for how the agent should interpret the user's 'free-text reply'.
- Capability inventory: The agent's broader capabilities (such as code execution or file modification) could be indirectly influenced by the 're-plan against the answers' instruction.
- Sanitization: None; the skill lacks mechanisms to sanitize or validate user input before it is used to 'fold it in' to the project plan.
Audit Metadata