MCP servers and skills: search, filter by where they run, and compare.
browser-use for Windows desktops — AI agents see and control any Windows app via UI Automation. The missing piece for computer-use agents.
SkillMIT★ 5No key
Windows computer-use MCP server: drive any desktop app via semantic discover-then-act targeting (entities + leases, not pixel coordinates), with per-action perception guards, a native Rust UIA engine,
Node.jsMIT★ 22No key
Multi-layer desktop automation for AI agents — browser CDP, Windows UIA, vision/OCR. Successor to AIWander/hands.
SkillApache-2.0★ 1Few starsNo key
Open-source Windows RPA and MCP server for AI agents — UIA desktop automation, browser DOM automation, ERP workflows, and fast element-level execution beyond screenshot-by-screenshot Computer Use.
PythonMIT★ 1Few starsNo key
Windows computer use for AI agents: host-independent Python CLI + JSON, Playwright, UIA/Win32 and optional Agent Skill.
SkillMITFew starsNo key
Safe browser and Windows computer-use tools for OpenCode, exposed through MCP.
Node.jsMIT★ 1Few starsNo key
Windows desktop automation via W3C WebDriver protocol — like ChromeDriver, but for Windows apps
SkillMIT★ 18No key
One vision layer for AI agents on Windows - a single MCP server to see and act on desktop apps, Electron windows, and the browser (undetected). Screen as structured text, so text-only LLMs work too.
PythonMITFew starsNo key
跨框架 Agent Skills 合集：视频笔记生成、本地识图、Windows 桌面/浏览器自动化，外加 Hermes 语言路由插件。DSH 插件已拆分为独立仓库。
SkillMIT★ 6No key
Local Windows control for Codex: webpage elements, desktop controls, screenshots and mouse/keyboard input. Optional Jev.
SkillMIT★ 2Few starsNo key
browser-use, but for the entire desktop — give any AI Agent eyes and hands on Windows. MCP + SDK, VLM-optional, with visual loop, introspection and meta-learning.
PythonMIT★ 3Few starsNo key
MCP server for computer use & browser automation - screenshot, OCR, click, type, find_text, Chrome/Electron CDP, template matching. macOS, Windows & Android. Works with Claude, Cursor, and any MCP cli
SkillMIT★ 139No key
ACI (Agent-Computer Interface) — The open standard that turns any application into a structured, agent-operable interface. One protocol. Any software. No screenshots.
SkillApache-2.0★ 7No key
Windows desktop automation MCP server - screenshots, OCR, mouse and keyboard, and accessibility trees via UI Automation, MSAA and Chrome DevTools Protocol. Node.js + built-in PowerShell, no Python and
Node.jsApache-2.0★ 3Few starsNo key
Secure MCP gateway that lets ChatGPT, Claude, Grok & other AI agents remotely code, run commands, manage Git, control browsers and desktop apps, operate multiple workspaces, and supervise your device/
Node.jsMIT★ 2Few starsNo key
Portable MCP runtime with 21 tools for processes, files, images, Chromium profiles and Windows desktop control. Shared OpenAI and Cloudflare connections with owner-approved OAuth.
GoAGPL-3.0★ 1Few starsNo key
Who stands behind it
