Uindow
OfficialRelated Servers
Alternatives to Uindow
- AlicenseAqualityCmaintenanceAgent-native browser control MCP server that enables AI agents to browse and interact with web pages via accessibility tree snapshots and ref ID-based commands.152MIT
Related Servers
- FlicenseNot gradedqualityDmaintenanceWindows GUI automation MCP server for Claude Code, enabling deterministic desktop app control via 32 tools including window management, input, pixel search, and file verification, leveraging the AutoIt backend.-
- FlicenseNot gradedqualityBmaintenancecivyk-winwright is a Playwright-style MCP server for Windows desktop and browser automation. It provides 59 tools for controlling WPF, WinForms, Win32 apps, and Chrome/Edge via the Model Context Protocol. A unified automation layer connecting Claude and MCP-compatible agents to native Windows UI and web interfaces.19-
- AlicenseNot gradedqualityCmaintenanceA Model Context Protocol server that gives AI assistants full control over native Windows applications — launch, inspect, click, type, screenshot, and test any WinUI3, WPF, WinForms, UWP, or Win32 app.44 npm39MIT
- AlicenseNot gradedqualityAmaintenanceA local MCP server that lets an assistant observe and control Windows application windows, plus interact with Xiaomi MiMo AI Desktop workflows, through 22 tools for Unicode typing, smooth dragging, cross-window drops, and bounded SSE streaming.MIT
- AlicenseAqualityAmaintenanceVerified control of the whole desktop for agents on Windows and Linux: apps with no API, windows, shell, browser. MCP server + CLI + REST API.3925MIT
- AlicenseNot gradedqualityAmaintenanceEnables AI clients to securely control and interact with a local Windows machine through 218 configurable tools for files, Git, processes, Windows UI, browser automation, WSL, Office, recovery, skills, and child MCP servers.5 npmMIT
TDQS
Scored across 38 tools
Tools are cleanly separated into app lifecycle, agent management/execution, and module-source editing groups. Potential lookalikes like agent_start/agent_execute and agent_status/agent_logs are explicitly distinguished in their descriptions. Even the many type-specific input/output setters are unambiguous because each name names the type.
Naming follows a predictable domain prefix plus snake_case action pattern (app_*, agent_*, src_*). Type-specific setters like src_input_set_bool and src_output_set_table are parallel and easy to generalize. No mixed casing or arbitrary verb styles appear.
38 tools is well above the 25+ threshold and feels heavy even for a broad automation/module-authoring server. The type-specific input and output setter tools (10 variants) inflate the count and could have been consolidated behind one generic setter with a type parameter. Still, no tool feels completely redundant.
The surface covers the full workflow: docs, app lifecycle, agent CRUD/status/execution/logging/screenshots, source export/import, FSM/function editing, keys management, and all input/output definitions. There are no obvious dead ends or missing lifecycle operations.