setup-rhdh-skills
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a helper script (scripts/setup.py) to execute system commands using subprocess.run with shell=False. This is primarily used to invoke the npx skills CLI for managing skill installations. The script includes a validation function (_operation_error) that restricts allowed commands to a specific whitelist (npx skills add), preventing arbitrary command injection.
- [EXTERNAL_DOWNLOADS]: The skill automates the installation of an entire suite of AI agent skills from a remote pack repository (redhat-developer/rhdh-skills). It also provides instructions for downloading and installing the gogcli tool from an external site (gogcli.sh) to support Google Workspace integrations.
- [SAFE]: The skill demonstrates several security-positive patterns: it enforces a 'mutation gate' requiring human approval of generated plans before execution; it instructs the agent to never read, print, or output sensitive credentials (API tokens, OAuth secrets) into the conversation; and it relies on standard system credential stores rather than creating insecure local files.
Audit Metadata