convez.ai
MCP ServerAutomaçãoAvançado
sandraschi

Windows Computer Use MCP

sandraschi/windows-computer-use-mcp

Automação agentica completa para Windows via MCP com 22 ferramentas e motor de missões

Ver no GitHub
BomPluginScore · de 0 a 100

Servidor MCP para automação agentica de computadores Windows. Fornece 22 ferramentas para cliques, digitação, screenshots, OCR e inspeção de UI, além de um motor de missões autônomo para executar tarefas complexas sem intervenção humana. Pode ser usado como servidor MCP, agente independente, aplicação web ou desktop.

Instalação

$npx convez-ai sandraschi/windows-computer-use-mcp

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
55
Forks
16
Acompanhando
0
Saúde
100%
Último commit
há 2 semanas
Versão
v0.8.0
Issues abertas
3
Contribuidores
3

Casos de uso

  • Automação de testes de UI em massa (100+ instaladores em uma execução)
  • Execução de workflows multi-aplicações (Notepad, Calc, Paint e qualquer app Windows)
  • Gravação e replay de macros com verificação autônoma de resultados
  • Inspeção e manipulação de elementos de UI via UIA/Win32

Veredito

Ferramenta robusta e production-ready para automação agentica em Windows. Oferece múltiplos modos de uso (MCP, agent, web, desktop) com logging completo e retry inteligente—ideal para cenários de escala com custo LLM mínimo.

README

A tradução para o português deste README ainda está na fila.

windows-computer-use-mcp

A tool for agents, and an agent itself.

YouIt
Use it as an MCP serverClaude, Cursor, DeepSeek call automation_click, automation_screenshot, automation_ocr — 22 tools
Use it as an autonomous agentGive it a goal: automation_mission(run="install app, verify UI, screenshot result") — it plans, executes, retries, and reports
Use it as a webappstart.ps1 opens a React dashboard at http://127.0.0.1:10788 with HITL, crawler, logging
Use it as a desktop appThe NSIS installer bundles everything into one binary — no Python, no uv, no git needed

Exhibit A: 100 Tauri/NSIS installers, one unattended run, $2 in LLM costs. Install, screenshot, verify, report — zero human intervention. That is what agentic Windows automation looks like at scale.

Built on pywinauto. Read docs/SAFETY.md before production use.


Quick Start

MethodCommand / Config
MCP stdio (Cursor, Claude Desktop){ "mcpServers": { "windows-computer-use": { "command": "uv", "args": ["--directory", "<PATH>", "run", "windows-computer-use-mcp"] } } }
HTTP streamable (any MCP HTTP client){ "mcpServers": { "windows-computer-use": { "url": "http://127.0.0.1:10789/mcp" } } }
Web operator UI.\start.ps1 → http://127.0.0.1:10788
Desktop app (NSIS installer)Download from Releases — zero deps

See INSTALL.md for detailed setup. Run just demo for examples.


Features

  • Window Management — find, activate, maximize, minimize, position, close
  • Mouse & Keyboard — click, drag, type, hotkeys, app shortcuts
  • UI Elements — inspect, click, read text, verify state via UIA / Win32
  • Visual Intelligence — screenshots, OCR, template matching
  • Autonomous Missions — give it a goal, it plans and executes with retry + verification
  • Macro Recording — record any UI sequence, replay, verify outcomes
  • Multi-App Workflows — chain actions across Notepad, Calc, Paint, or any Windows app
  • Telemetry — every action logged to SQLite; query failure patterns by tool
  • Adaptive Location — auto-cascades through title/auto_id/control_id/class/OCR to find elements
  • Face Recognition — optional, off by default

Documentation

DocContent
INSTALL.mdSetup: desktop app, uv, MCP config
docs/README.mdFull documentation hub
docs/py-stack.mdPython dependency deep dive
docs/composing-with-playwright.mdBrowser automation with Playwright MCP
docs/ocr.mdOCR system — Tesseract setup, limitations, competition
docs/cua-nsis-certification.mdDogfooding: using the tool to test its own NSIS installer
docs/ROADMAP.mdImprovement roadmap short/medium/long term
docs/SAFETY.mdHITL, kill switch, opt-in features
docs/TOOLS.mdPortmanteau tool reference
tests/README.mdTest suite guide and e2e setup
examples/README.mdRunnable demos
mcpb/README.mdMCPB bundle packaging
web_sota/README.mdOperator UI build/dev guide
CHANGELOG.mdRelease history

Ports

PortService
10788Frontend — Vite operator UI
10789Backend — FastAPI + FastMCP HTTP
stdioMCP transport (port-free)

Related

RepoWhat it does
autohotkey-mcpRaw input recording/replay via AHK
browser-mcpPlaywright browser control — for webapps, HTML DOM, websites
virtualization-mcpSandbox / VM isolation
windows-operations-mcpRegistry, services, accounts

Browser vs desktop: This server drives Win32 / UI Automation. For HTML/DOM and websites, pair with browser-mcp (Playwright). Both MCPs can run side by side — use one profile that loads both and let the LLM pick the right tool for the target.

Fleet standards: mcp-central-docs.


License

MIT — Copyright (c) 2026 Sandra Schipal.

Ver o original no GitHub

Bom para

Relacionados