using-flywheel
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands to verify the presence of skills in specific local directories (
~/.claude/skills/validate, etc.). This is a standard check for verifying local environment configuration. - [EXTERNAL_DOWNLOADS]: The documentation references an external website (
agent-flywheel.com) for a multi-agent factory stack and mentions a "single-curl install" on a dedicated VPS. While this describes a remote installation process, it is presented as a prerequisite setup step for the user to perform on their own infrastructure, not as a hidden or automated execution by the agent itself. - [COMMAND_EXECUTION]: The
scripts/validate.shfile executes a local maintenance script (heal.sh) from the repository's toolchain usingexec. This is a standard validation pattern for ensuring skill integrity within this specific ecosystem.
Audit Metadata