interview
Pass
Audited by Gen Agent Trust Hub on May 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill facilitates a structured interview process to generate project specifications. All operations, including reading plan files and writing specifications, align with the stated purpose and use appropriate platform tools.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes untrusted content from user-provided plan files without explicit boundary markers or instructions to ignore embedded commands. However, the structured nature of the interview logic (analyzing specific categories like architecture and UX) mitigates the risk of the agent deviating from its primary task.
- Ingestion points: Reads user-provided plan files from paths passed via $ARGUMENTS (SKILL.md).
- Boundary markers: None present; the agent is instructed to read and analyze the plan file directly.
- Capability inventory: Uses Read, Write, and AskUserQuestion tools.
- Sanitization: No input sanitization or validation is performed on the content of the plan file.
Audit Metadata