skills/timblazing/skills/shadcn/Gen Agent Trust Hub

shadcn

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill correctly restricts its capabilities to the official shadcn CLI (npx shadcn@latest) via the allowed-tools frontmatter, preventing arbitrary command execution.
  • [SAFE]: Dynamic context injection (!npx shadcn@latest info --json) is used legitimately to retrieve project-specific configuration (aliases, Tailwind version, installed components) without accessing sensitive user credentials or environment files.
  • [SAFE]: The instructions explicitly forbid manual file fetching from untrusted sources, directing the agent to use the CLI's built-in resolution and diffing mechanisms (--dry-run, --diff, --view) for secure component updates.
  • [SAFE]: External references to ui.shadcn.com and GitHub repositories are documented as standard parts of the shadcn ecosystem for fetching documentation and registry metadata.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 05:54 PM