The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The usage examples in examples/usage.md recommend the installation and use of a package named tauri-plugin-plugin-permissions. This package is not part of the official @tauri-apps namespace, nor does it follow the author's verified naming patterns, making it an unverifiable dependency that could lead to the execution of untrusted code if installed.
[PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing user-provided functional requirements to generate capability configurations.
Ingestion points: User-provided functional requirements and feature descriptions for plugin permissions used during the mapping and generation steps described in SKILL.md.
Boundary markers: The instructions do not define delimiters or provide warnings for the agent to ignore instructions embedded within the user data.
Capability inventory: The skill facilitates the generation of capabilities/default.json content, which defines sensitive application permissions such as fs:read-text-file and http:default (as referenced in templates/capabilities.json).
Sanitization: The skill lacks logic to validate or sanitize user-provided feature names or descriptions before they are incorporated into the generated permission schemas.

tauri-app-plugin-permissions