command-skill-creator
Pass
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional and provides templates for building multi-phase workflows. It does not perform any sensitive operations or network requests itself.\n- [SAFE]: The instructions explicitly guide users toward security best practices, such as validating arguments, avoiding hardcoded absolute paths, and performing discovery via relative glob patterns to prevent context leakage or blind modifications.\n- [SAFE]: It emphasizes the 'disable-model-invocation: true' setting for all commands with side effects, which is a key security control within the platform to prevent the AI from autonomously executing potentially dangerous operations.
Audit Metadata