The Agent Skills Directory

[Unverifiable Dependencies] (MEDIUM): The skill provides instructions to install the uview-pro package using npm, yarn, and pnpm. While this is a common procedure for library documentation, the package is not hosted in a repository or organization identified as a 'Trusted External Source' (such as Google, Microsoft, or Vercel). The integrity and security of the package must be verified independently before use.
[Indirect Prompt Injection] (LOW): The skill defines a significant attack surface by instructing the agent to ingest content from various local directories (examples/, api/, templates/) to fulfill user requests.
Ingestion points: Files located in examples/components/, examples/tools/, api/, and templates/ folders.
Boundary markers: Absent. The skill does not define specific delimiters or warnings to ignore instructions embedded within the loaded documentation or templates.
Capability inventory: No active capabilities such as subprocess execution, file writing, or network operations are present in the provided skill manifest. The risk is currently limited to influencing agent reasoning or output formatting.
Sanitization: Absent. There is no evidence of validation or filtering for content retrieved from the referenced directories.

uview-pro-vue3