groundwork-session
Pass
Audited by Gen Agent Trust Hub on Aug 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill exhibits no malicious behavior. It performs file read and write operations within a specified data folder, which are necessary for maintaining coaching session history and user profiles.
- [PROMPT_INJECTION]: The skill processes user input and external framework files to conduct sessions. While this constitutes a standard attack surface for interactive agents, the skill follows a strict coaching logic that focuses on specific session goals and structured moves.
- Ingestion points: User chat input, profile.md, and framework markdown files.
- Boundary markers: None explicitly defined in the instructions for isolating external file content.
- Capability inventory: Limited to file I/O within the data folder and invocation of the groundwork-synthesize skill.
- Sanitization: None described for external content or user responses.
Audit Metadata