skill-authoring
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): The skill consists entirely of documentation and guidelines for skill development. No malicious code, prompt injections, or obfuscation techniques were detected.
- [COMMAND_EXECUTION] (SAFE): While the documentation references the use of local validation scripts (e.g.,
npm run references,npm run tokens), these are standard developer workflows for ensuring repository compliance and do not involve untrusted remote execution or elevated privileges. - [DATA_EXFILTRATION] (SAFE): There are no patterns indicating unauthorized data access or exfiltration. Network references are limited to trusted documentation sites like GitHub and agentskills.io.
- [PROMPT_INJECTION] (SAFE): The instructions are clear, action-oriented, and follow the intended purpose of assisting developers. No attempts to bypass AI safety filters or override system instructions were found.
Audit Metadata