interview-analyst
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection through the ingestion of external data. Ingestion points: Documents located in the /theory and /interviews directories. Boundary markers: None; the instructions do not specify the use of delimiters or 'ignore' instructions for the analyzed text. Capability inventory: The skill has the ability to write multiple markdown files to the /analysis directory and mentions the use of web search for literature in Phase 3. Sanitization: No input validation or escaping is applied to the ingested documents before they are processed by the agent.
Audit Metadata