behavioral-modes
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill contains only instructional documentation and behavioral prompts with no code or command execution capabilities.
- [DATA_EXFILTRATION]: Functional tool access is restricted to local file reading, and there are no instructions to transmit data externally.
- [PROMPT_INJECTION]: Instructions focus on task-specific heuristics and do not attempt to override system safety rules or extract internal prompts.
- [REMOTE_CODE_EXECUTION]: No external URLs or package managers are referenced for downloading or executing code.
Audit Metadata