web-design-expert

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses WebFetch and specific MCP tools like mcp__magic__21st_magic_component_inspiration and mcp__magic__logo_search to retrieve external content including UI patterns, component designs, and brand logos from 21st.dev and other web sources.
  • [PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by ingesting and processing untrusted data from external web sources and component registries.
  • Ingestion points: Data enters the context through WebFetch (SKILL.md) and component inspiration/logo search tools (references/tooling-integration.md).
  • Boundary markers: The instructions do not define clear delimiters or 'ignore' instructions for content retrieved from external UI component searches.
  • Capability inventory: The skill has the ability to Write and Edit files, and generates production-ready code using mcp__magic__21st_magic_component_builder based on retrieved inspiration.
  • Sanitization: There is no evidence of sanitization or schema validation for the React/Tailwind code snippets or metadata fetched from external tools before they are integrated into the project workspace.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 11:54 AM