skills/hairyf/skills/nuxt/Gen Agent Trust Hub

nuxt

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The documentation for Nuxt Layers (advanced-layers.md) describes the framework's ability to fetch and extend applications using remote Git repositories from GitHub, GitLab, and Bitbucket.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents server-side utilities (features-server.md) for ingesting external data via request bodies (readBody), query parameters (getQuery), and cookies (getCookie), which establishes an ingestion surface for untrusted data if handled by an agent without proper sanitization.
  • [DYNAMIC_EXECUTION]: The module authoring guide (advanced-module-authoring.md) describes build-time code generation utilities such as addPluginTemplate, which creates executable virtual files from provided string templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 08:05 PM