foreman
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a set of instructions for a hierarchical agent collaboration model ("Foreman"). It focuses on decomposition, planning, and verification of code changes by the main agent.
- [SAFE]: No sensitive file access, network exfiltration, or hardcoded credentials were found. The skill operates within the local repository context using existing tools.
- [SAFE]: There are no obfuscated commands, Base64 strings, or hidden URLs detected in the files.
- [SAFE]: The workflow explicitly includes human-in-the-loop or Foreman-in-the-loop verification steps, requiring the inspection of actual diffs ("Inspect the diff, not the report") rather than just agent summaries, which is a security-positive behavior for delegated tasks.
Audit Metadata