model-reference
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The skill provides pre-defined ripgrep (
rg) command templates to assist the agent in searching the localreferences/directory for model metadata. These commands are static and intended for information retrieval within the skill's internal files. - [EXTERNAL_DOWNLOADS]: The reference files contain links to official documentation, repositories, and model cards for well-known AI organizations (e.g., Anthropic, Google, Meta, Microsoft, OpenAI). These URLs point to official, trusted service domains for technical reference purposes.
- [SAFE]: No suspicious patterns such as prompt injection, base64-encoded payloads, or unauthorized data access were found. The skill operates as a static reference for model specifications and follows best practices by directing the agent to official sources.
Audit Metadata