skills/onmax/nuxt-skills/nuxt-ui/Gen Agent Trust Hub

nuxt-ui

Pass

Audited by Gen Agent Trust Hub on Nov 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to configure an external MCP server at https://ui.nuxt.com/mcp. This server provides the AI agent with real-time access to component documentation, props, and examples for the Nuxt UI library.
  • [INDIRECT_PROMPT_INJECTION]: The references/layouts/chat.md file provides patterns for building AI chat interfaces, which naturally ingest untrusted data from external sources and streaming LLM responses.
  • Ingestion points: Data enters the application context via the readBody(event) function in the server-side API handler and the useChat composable in the client-side Vue files.
  • Boundary markers: Content is rendered using the Comark component, which properly segregates markdown tokens and facilitates the use of Nuxt UI prose components.
  • Capability inventory: The provided chat patterns include support for tool calls (isToolUIPart), which the agent can use to perform actions based on model instructions.
  • Sanitization: Content rendering relies on the Comark library for markdown parsing and standard Nuxt UI components for structured data display, following industry-standard safety practices for AI-enabled UIs.
  • [SAFE]: All suggested dependencies and remote resources originate from well-known technology providers or the official project documentation site associated with the skill's primary purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 15, 2026, 04:49 PM