The Agent Skills Directory

[SAFE]: Reference to official Tauri 2.0 plugins. The skill documentation covers standard packages like @tauri-apps/plugin-fs and @tauri-apps/plugin-http, which are official components of the Tauri ecosystem.
[SAFE]: Security-centric instruction. The skill explicitly warns against broad shell permissions and provides examples of using validators to restrict command execution, adhering to the principle of least privilege.
[INDIRECT_PROMPT_INJECTION]: The skill provides patterns for ingesting data that could contain indirect instructions. 1. Ingestion points: readTextFile in Pattern 1, fetch in Pattern 3, and readText (clipboard) in Pattern 6. 2. Boundary markers: The instructions focus on Tauri's technical boundary markers, such as BaseDirectory and capability-based permission scoping. 3. Capability inventory: The skill documents capabilities for executing shell commands (Command.execute), writing to the file system (writeTextFile), and making network requests (fetch). 4. Sanitization: The documentation promotes the use of regex validators within the shell:allow-execute permission scope to sanitize command arguments.

tauri-syntax-plugins-api