agents-md-generator
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs read-only analysis of the repository using standard command-line tools like
tokei,rg(ripgrep), andgit. These operations are used solely to gather facts for documentation generation. - [SAFE]: Python scripts bundled with the skill (
loc_to_limit.py,detect_monorepo.py,git_ownership_signals.py) use thesubprocessmodule to invoke system utilities with list-based arguments, avoiding shell injection risks. - [SAFE]: The skill implements a 'management gate' and ownership checks to ensure it only updates files it previously generated, protecting externally-authored documentation from accidental modification.
- [SAFE]: There is no evidence of network exfiltration, credential harvesting, or persistence mechanisms. The output is strictly confined to local Markdown file creation/modification.
- [SAFE]: The skill includes explicit scope boundaries, such as prohibiting the spawning of subagents to maintain a unified and secure execution context.
Audit Metadata