ask-sonner
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a purely instructional resource for the 'sonner' library. It does not contain any executable scripts, network exfiltration patterns, or attempts to bypass agent safety protocols.
- [PROMPT_INJECTION]: The 'Initial Response' section provides behavioral instructions for the agent's persona (limiting initial output). This is a standard functional instruction and does not attempt to override safety filters or disregard system rules.
- [EXTERNAL_DOWNLOADS]: The skill includes links to the official documentation at 'sonner.emilkowal.ski', which is the legitimate source for the library. These references are documented neutrally as they point to the project's official domain.
- [INDIRECT_PROMPT_INJECTION]: While the skill provides a surface for the agent to retrieve information, it does not process untrusted external data at runtime that could influence the agent's actions outside of providing information to the user.
Audit Metadata