shadcn-first
Warn
Audited by Socket on Sep 14, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The core purpose is coherent, and the official shadcn CLI/MCP usage is broadly consistent with a UI skill. The main risk is disproportionate trust in non-official third-party infrastructure (`@jpisnice/shadcn-ui-mcp-server` and prompt-kit registry), plus unpinned `@latest` execution and remote code import into the repo. No confirmed malware or explicit credential theft is visible, but the install/data-flow footprint exceeds what is necessary for a strictly official shadcn workflow.
Confidence: 84%Severity: 74%
Audit Metadata