Agent Skills: tauri-app-shell

When to use this skill

ALWAYS use this skill when the user mentions:

• Executing system commands from a Tauri app
• Shell plugin permissions and command allowlists
• Opening external URLs via shell (alternative to opener plugin)

Trigger phrases include:

• "shell", "execute command", "spawn process", "allow-execute", "system command"

How to use this skill

1. Install the shell plugin:

   cargo add tauri-plugin-shell
   

2. Register the plugin in your Tauri builder:

   tauri::Builder::default()
       .plugin(tauri_plugin_shell::init())
   

3. Configure strict allowlists in src-tauri/capabilities/default.json:

   {
     "permissions": [
       {
         "identifier": "shell:allow-execute",
         "allow": [{ "name": "git", "cmd": "git", "args": ["status"] }]
       },
       "shell:allow-open"
     ]
   }
   

4. Execute commands from the frontend:

   import { Command } from '@tauri-apps/plugin-shell';
   const output = await Command.create('git', ['status']).execute();
   console.log('stdout:', output.stdout);
   

5. CRITICAL: Apply strict allowlists -- the shell plugin is high-risk; never allow arbitrary command execution
6. Use regex constraints on arguments when dynamic input is needed, and validate all user-provided args

Outputs

• Shell plugin setup with strict command allowlists
• Command execution pattern with output handling
• Security-first configuration (this is a high-risk plugin)

References

• https://v2.tauri.app/plugin/shell/

Keywords

tauri shell, execute command, spawn process, allow-execute, system command, security