Skip to content

tauri-agent-tools

Agent-driven inspection toolkit for Tauri desktop apps.

25 commands to screenshot, inspect, interact with, and monitor Tauri apps from the CLI.

License: MIT Node >= 20

tauri-agent-tools

The Problem

Debugging frontend issues in Tauri desktop apps requires manually screenshotting, cropping, and describing what you see. Existing tools either hijack your cursor (xcap-based), render DOM to canvas (html2canvas — can't capture WebGL/video/canvas elements), or have no authentication.

The Solution

Combine a bridge's knowledge of element positions (getBoundingClientRect) with real pixel screenshots (import -window + ImageMagick crop). No other tool does this.

Features

  • DOM-targeted capture — screenshot any CSS-selectable element with real screen pixels
  • 25 commands — 14 read-only inspection commands plus 11 interaction and workflow commands (click, type, scroll, focus, navigate, select, invoke, probe, capture, check, store-inspect)
  • Cross-platform — Linux X11, Linux Wayland/Sway, Linux Wayland/Hyprland, macOS
  • Token authenticated — random 32-char token, localhost-only bridge
  • Agent Skills — ships agentskills.io skills for AI coding agents
  • JSON output — all commands support --json for structured output

Quick Install

npm install -g tauri-agent-tools

See Installation for platform-specific requirements.

Quick Example

# Screenshot a specific DOM element
tauri-agent-tools screenshot --selector ".toolbar" -o /tmp/toolbar.png

# Explore the DOM tree
tauri-agent-tools dom --depth 3

# Monitor IPC calls in real-time
tauri-agent-tools ipc-monitor --duration 10000

Next Steps