conducting-user-interviews
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- PROMPT_INJECTION (LOW): The skill identifies a workflow for ingesting and synthesizing external interview data, creating a surface for indirect prompt injection. Ingestion points: Verbatim interview notes and transcripts processed in SKILL.md (Workflow Step 5-7) and references/WORKFLOW.md. Boundary markers: Absent (no explicit delimiters for the processed verbatim text). Capability inventory: None (no subprocess execution, file system modification, or network requests found in the skill). Sanitization: Absent.
- NO_CODE (SAFE): The skill consists entirely of Markdown documentation and one metadata JSON file; no executable scripts or code components are present.
Audit Metadata