tool-specification
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely instructional, providing a methodology for defining AI agent tool specifications. It does not include any scripts, commands, or remote dependencies.
- [SAFE]: The content promotes security best practices, such as implementing a three-tier permission gating model (Prohibited, Explicit Permission, and Regular) to mitigate risks associated with tool use.
- [NO_CODE]: There are no code snippets, shell commands, or executable scripts provided in the skill file, reducing the attack surface to zero.
Audit Metadata