conducting-deep-interviews
Pass
Audited by Gen Agent Trust Hub on Jul 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a complex instructional framework for guiding users through a requirements-gathering interview. It does not include executable code, network operations, or credential access.
- [EXTERNAL_DOWNLOADS]: The
references/harness-source-notes.mdfile lists several GitHub repositories as design references (e.g., github.com/Q00/ouroboros, github.com/mattpocock/skills). These are cited for research purposes and the skill explicitly notes they are not downloaded or executed at runtime. - [COMMAND_EXECUTION]: The instructions direct the agent to read repository files and documentation to provide context for the interview. This behavior is standard for development-focused agents and is restricted to the local project environment.
- [DATA_EXFILTRATION]: No network calls or data exfiltration patterns were identified. The skill includes an optional file-write step for the final brief, which defaults to the local project root and requires user authorization.
Audit Metadata