model-architecture-diagram
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references and provides instructions for downloading a model architecture gallery from the author's GitHub repository. It also indexes image links from well-known repositories such as Tencent, Alibaba, and DataWhale. These references are used for displaying diagrams and are sourced from established services.
- [COMMAND_EXECUTION]: The skill executes a bundled Python script to resolve model names. The script is self-contained, uses standard Python libraries, and operates on a static JSON index file included in the skill.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied model names. The resolver script implements input sanitization using NFKC normalization and character filtering (allowing only alphanumeric characters, dots, dashes, and Chinese characters) to ensure the input is safe for use in queries and does not cause unexpected behavior.
Audit Metadata