shape
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill performs design planning by conducting user interviews and referencing local project documentation (
.design-context.md,CLAUDE.md). It operates entirely within the expected project scope and follows standard interaction patterns for design discovery. - [SAFE]: The skill recommends other specialized design tools and services (e.g., shadcn, fal.ai, ImageMagick) for implementation, but does not execute remote code or perform unauthorized network operations itself. These references are informational and aid in design-to-code routing.
- [SAFE]: Access to local configuration files (e.g., aesthetic principles in
~/.claude/skills/) is used appropriately to inform design decisions based on established guidelines and does not involve data exfiltration or credential harvesting.
Audit Metadata