skills/21st-dev/magic-mcp/21st-ui/Gen Agent Trust Hub

21st-ui

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill facilitates the retrieval of component code and SVG logos from the vendor's official domain at 21st.dev. This behavior is the primary intended function of the skill and originates from the author's own infrastructure.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data (UI code, descriptions, and metadata) returned from API calls. While this presents an ingestion surface for third-party content, it is inherent to the marketplace workflow.
  • Ingestion points: Data retrieved via search, get_component, and generate tools in SKILL.md.
  • Boundary markers: No specific delimiters or safety instructions are defined in the skill file for the integrated code.
  • Capability inventory: The skill is designed to integrate code into the project, which may involve file system writes.
  • Sanitization: The instructions do not specify content validation or sanitization routines.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 05:59 PM