mcp-window-vision

# MCP Window Vision A small, **cross-platform** [MCP](https://modelcontextprotocol.io) server (macOS / Windows / Linux) that gives an AI assistant **eyes on your screen** and access to **Unity's Console log** — so an agent working on a GUI app (like a Unity project) can *see* the result and *read* the errors itself, instead of asking you to screenshot and paste. Screenshots use [`mss`](https://pypi.org/project/mss/) + [`Pillow`](https://pypi.org/project/Pillow/) (both cross-platform). The MCP transport is pure standard library. Runs on Python 3.9+. ## Tools | Tool | What it does | |------|--------------| | `capture_screen` | Screenshot the main display (downscaled), returned as a PNG image | | `capture_region` | Screenshot a rectangle `{x, y, w, h}` (pixels) | | `read_unity_console` | Tail Unity's `Editor.log` (Debug.Log / errors / compile messages); optional substring filter | ## Requirements - macOS, Windows, or Linux (Linux capture needs X11) - Python 3.9+ - macOS only: **Screen Recording permission** for whatever process launches the server (your terminal / MCP client): *System Settings → Privacy & Security → Screen Recording*. Unity `Editor.log` is found automatically per-OS: | OS | Path | |----|------| | macOS | `~/Library/Logs/Unity/Editor.log` | | Windows | `%LOCALAPPDATA%\Unity\Editor\Editor.log` | | Linux | `~/.config/unity3d/Editor.log` | ## Install ```bash pip install mcp-window-vision ``` This installs the `mcp-window-vision` console command. Verify screen capture works (and that permission is granted): ```bash mcp-window-vision --selftest # -> "capture OK — NNNNN base64 chars" ``` ## Use with Claude Code User scope (available in every project): ```bash claude mcp add window-vision --scope user -- mcp-window-vision ``` …or per-project via a `.mcp.json` at the project root: ```json { "mcpServers": { "window-vision": { "type": "stdio", "command": "mcp-window-vision" } } } ``` Restart the session so the client …