shadcn
Pass
Audited by Gen Agent Trust Hub on May 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of the shadcn CLI via project package runners (npx, pnpm dlx, bunx) to perform project audits, search registries, and manage component lifecycles. This is the primary and intended mechanism for the skill's functionality.- [EXTERNAL_DOWNLOADS]: The skill is configured to fetch component definitions, metadata, documentation, and code examples from the shadcn registry (ui.shadcn.com) and other community-provided registries or URLs.- [DYNAMIC_CONTEXT_INJECTION]: The main SKILL.md file uses the dynamic injection pattern (
!npx shadcn@latest info --json) to automatically refresh the agent's understanding of the local project configuration, including installed components and framework versions, whenever the skill is loaded.- [REMOTE_CODE_EXECUTION]: The CLI's 'add' and 'apply' commands facilitate the download and integration of React component source code from remote registries directly into the local project structure. To mitigate risks from third-party code, the skill explicitly instructs the agent to review and verify all newly added files.- [PROMPT_INJECTION]: The skill includes instructions to fetch and process external documentation and example URLs. While this represents a potential surface for indirect prompt injection, the skill incorporates safety guidelines requiring the agent to audit downloaded content before finalizing changes.
Audit Metadata