The Agent Skills Directory

[METADATA_POISONING]: The skill metadata in SKILL.md claims the author is github.com/vuejs-ai, which contradicts the actual author rockcookies. This impersonation of a well-known community organization is deceptive and could lead users to trust the content based on a false origin.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process and refactor Vue.js project files, including .vue, .js, and .ts files. This creates a vulnerability where malicious instructions embedded in a project's source code could influence the agent's behavior during analysis.
Ingestion points: The skill explicitly loads for any .vue files, Vue Router, Pinia, or Vite-related work as stated in SKILL.md.
Boundary markers: There are no instructions or delimiters defined to help the agent distinguish between its core instructions and potentially malicious content within the project files it reads.
Capability inventory: The skill examples and documentation utilize network requests (fetch in references/updated-hook-performance.md) and state management updates, which could be abused if an injection occurs.
Sanitization: While the skill correctly recommends DOMPurify for sanitizing HTML rendered in templates, it lacks similar sanitization or validation for the code/data it ingests into the agent context.

vue-best-practices