The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a configuration template in templates/capabilities.json that enables the execution of a Node.js sidecar binary (binaries/my-node-app) using the shell:allow-execute permission. This is the standard mechanism in Tauri v2 for running external binaries and is the primary purpose of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates an architecture where the application frontend communicates with a Node.js sidecar, creating an attack surface for data-driven instructions.
Ingestion points: Data is ingested from the frontend via invoke calls to plugin:sidecar-nodejs and emit events, as shown in examples/usage.md.
Boundary markers: No specific boundary markers or delimiters for untrusted input are provided in the configuration templates.
Capability inventory: The skill enables shell execution of the sidecar binary and binary argument passing.
Sanitization: The usage examples do not include explicit sanitization or validation logic for the data passed to the sidecar.
[EXTERNAL_DOWNLOADS]: The skill's documentation in examples/usage.md references the installation and use of tauri-plugin-sidecar-nodejs and @tauri-apps/api. The @tauri-apps/api package is a well-known service from the Tauri organization, and the sidecar plugin is presented as a component of the integration being taught.

tauri-app-sidecar-nodejs