Skip to content

Quick Start

There are two ways to use tauri-agent-tools: standalone (no bridge needed) and bridge-connected (full feature set).

Standalone (No Bridge)

These commands work with any window — no Tauri bridge required.

1. List windows

tauri-agent-tools list-windows

This shows all visible windows with their IDs, PIDs, and sizes. Tauri apps with an active bridge are marked.

2. Get window info

tauri-agent-tools info --title "My App"

Shows window geometry, position, and display server info.

3. Full window screenshot

tauri-agent-tools screenshot --title "My App" -o /tmp/full.png

Captures the entire window as a PNG.

4. Wait for a window

tauri-agent-tools wait --title "My App" --timeout 10000

Polls until a window with the matching title appears.

Bridge-Connected (Full Features)

With the bridge set up in your Tauri app, you get access to DOM-targeted commands.

1. Check page state

tauri-agent-tools page-state

Shows URL, title, viewport size, scroll position, and Tauri detection.

2. Explore the DOM

tauri-agent-tools dom --depth 3
tauri-agent-tools dom ".sidebar" --depth 2 --styles

3. Screenshot a specific element

tauri-agent-tools screenshot --selector ".toolbar" -o /tmp/toolbar.png
tauri-agent-tools screenshot --selector "#main-canvas" --max-width 800 -o /tmp/canvas.png

4. Monitor IPC calls

tauri-agent-tools ipc-monitor --duration 10000

Watches Tauri IPC calls in real-time for 10 seconds.

5. Inspect storage

tauri-agent-tools storage --type local

6. Evaluate JavaScript

tauri-agent-tools eval "document.querySelectorAll('.item').length"

What's Next

Reliable automation

Select an app with --pid when several bridges are running. Use --window-label for a secondary webview and --window-id for an OS window. Title matching is regex on X11 and substring matching on macOS/Wayland.

tauri-agent-tools eval 'Promise.resolve(document.readyState)' --json
tauri-agent-tools check --selector '#app' --no-errors --duration 1000 --json
tauri-agent-tools capture -o /tmp/evidence --json

Check exit status and capture partial/warnings before treating evidence as complete. JSON results go to stdout; fatal error envelopes and warnings go to stderr.