The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's environment and workflows involve the execution of development tools including the Tauri CLI (tauri dev, tauri build) and Cargo to manage application lifecycle and signing.
[EXTERNAL_DOWNLOADS]: The skill references official resources at v2.tauri.app and incorporates provenance from established open-source repositories for agent skills and Tauri utilities.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted Tauri project configurations and source code, creating an attack surface for indirect prompt injection.
Ingestion points: The agent is instructed to read project files such as tauri.conf.json, Cargo.toml, and capability manifests in the src-tauri directory to understand project scope.
Boundary markers: The instructions (Rule 16) explicitly require the agent to prioritize path canonicalization and Rust-side validation to prevent frontend-driven exploits.
Capability inventory: The agent utilizes platform tools to build, check, and sign applications.
Sanitization: The skill provides specific guidance to avoid dangerous primitives like run_shell or read_any_file that could bridge the webview boundary to the operating system.

tauri