skills/metamask/skills/ui-development/Gen Agent Trust Hub

ui-development

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides standard design system alignment workflows and UI development guidelines for React and React Native environments. It encourages the use of established design system primitives over custom implementations.
  • [SAFE]: All referenced URLs point to legitimate vendor resources, including the official MetaMask design system repository on GitHub and an internal strategy Google Doc.
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest documentation from an external MCP server (storybook-broker-mcp) and local source files in node_modules. While this is a standard workflow for developer tools, it represents a surface where malicious instructions could potentially be introduced via external data.
  • Ingestion points: MCP tool storybook-broker-mcp, local package type definitions in node_modules (e.g., @metamask/design-system-react-native).
  • Boundary markers: Not implemented in the instructions.
  • Capability inventory: The skill performs file system reads and executes MCP tools to retrieve documentation.
  • Sanitization: No specific sanitization or filtering of the external documentation content is defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 07:59 PM
Security Audit — agent-trust-hub — ui-development