The Agent Skills Directory

[SAFE]: The skill provides legitimate technical documentation and workflow instructions for Vue 3 development. It follows industry best practices by recommending environment variables (e.g., process.env.SENTRY_AUTH_TOKEN) for sensitive configuration rather than hardcoding credentials.
[EXTERNAL_DOWNLOADS]: The skill references standard, widely-used packages and tools in the Vue ecosystem (e.g., Pinia, Nuxt, Quasar, Vite plugins). All suggested installations use official registries and established packages.
[COMMAND_EXECUTION]: Instructions include the use of standard development CLI tools such as npm, npx, and vue-tsc for project initialization, building, and type checking, which are expected in a professional developer workflow.
[INDIRECT_PROMPT_INJECTION]: As a coding assistant, the skill ingests user-provided code and requirements, creating an attack surface for indirect prompt injection, although no specific exploits are present.
Ingestion points: User-provided Vue components, project configurations, and application requirements (SKILL.md).
Boundary markers: None explicitly defined to separate developer instructions from potentially untrusted project data.
Capability inventory: The skill uses terminal-based validation and build commands (vite build, vitest, vue-tsc) as part of the core workflow (SKILL.md).
Sanitization: No specific input sanitization or validation routines for external code projects are described.

vue-expert