caveman-explore
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is configured for read-only access, explicitly forbidding command execution and file modification in both instructions and included automated tests.
- [SAFE]: Tool access is restricted to searching and reading (Read, Glob, Grep) via platform configuration in the frontmatter.
- [SAFE]: No external dependencies, network operations, or hardcoded credentials were identified in the source or tests.
- [SAFE]: The instructions mandate a strict output format (compact line citations), which serves as a natural constraint against large-scale data leakage into the solver agent's context.
Audit Metadata