herdr
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
subprocess.runwithincapture-cli-help.pyto automate the discovery of CLI help text. This is a legitimate diagnostic function used for the skill's documented maintenance procedures to ensure local instructions remain synchronized with the installed binary version. - [COMMAND_EXECUTION]: The skill instructions in
SKILL.mdandCLI-REFERENCE.mdguide the agent to execute shell commands via theherdrbinary. These operations (managing terminal panes, layouts, and background processes) are the primary intended purpose of the skill. - [SAFE]: External references to the
herdrdev/herdrGitHub repository are used solely for documentation and version pinning during maintenance reviews. No remote code is downloaded or executed at runtime. - [SAFE]: The
MAINTENANCE.mdfile defines a rigorous human-in-the-loop review procedure for upgrading the skill, which serves as a security best practice for managing tool-based skills.
Audit Metadata