create-design-md
Pass
Audited by Gen Agent Trust Hub on Jul 29, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill defines a structured process for an AI agent to analyze UI code, theme configurations, and component libraries to document a project's design language.
- [SAFE]: File operations are restricted to reading project source files and writing design documentation (DESIGN.md) or updating project configuration files (CLAUDE.md, AGENTS.md) with pointers to the design spec.
- [SAFE]: No network activity, external downloads, or execution of untrusted scripts were detected.
- [SAFE]: The skill uses clear, instructional language and does not employ obfuscation or prompt injection techniques.
- [SAFE]: Although the skill processes untrusted project data (UI code and styles) which presents an indirect prompt injection surface, the risk is negligible as the agent's capabilities are limited to generating descriptive markdown documentation.
Audit Metadata