Log inSubmit server
Open Compute (computer Use) logo

Open Compute (computer Use)

Local stdio MCP tools for observing and controlling an interactive Windows desktop with UIA targeting and safety gates.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Open Compute (computer Use)

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "open-compute-computer-use": {
      "command": "npx",
      "args": [
        "-y",
        "open-compute-mcp"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

We couldn’t confirm it starts correctly

We ran the install command but it didn't respond within our test window — this can mean a slow first-time install rather than a real problem.

npx -y open-compute-mcp

Process exited before answering initialize.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 6d ago.

Overview

The open-compute (computer-use) MCP server exposes model-agnostic desktop observation and interaction tools through local MCP stdio. It connects an MCP client to an interactive desktop session, using screenshots, Windows UI Automation elements, window tokens, and safety-gated actions. Use it for agents that need to inspect or operate Windows applications rather than call a remote API. Real capture and input must run on the MCP client’s own host, so hosted deployment is not suitable.

Use cases

  • Inspect Windows application screens before taking action
  • Target desktop controls through Windows UI Automation
  • Run safety-gated mouse and keyboard workflows
  • Monitor directories and desktop session status
  • Request human screen feedback or voice notes

Key features

  • Screenshot capture with observation IDs
  • Windows UIA tree inspection and semantic targeting
  • Safety-gated coordinate and UIA actions
  • Window discovery with issued tokens
  • Desktop signal overlays and abort handling
  • Directory watching and voice-note capture

Tools16

What Open Compute (computer Use) lets your agent do. Expand a tool for its parameters.

Self-reported~317 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked Today

Recent health check succeeded.

8/8 checks healthy over the last 45d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 23h ago via OSV.dev · open-compute-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars1Stargazers on the source repository.
npm downloads216Package downloads in the last 30 days.
Last commit15d agoMost recent push to the default branch.
Install checkInconclusiveDidn't respond in our test window — often a slow first install.
Tools exposed16Callable tools this server registers over MCP.
Directory activity9 viewsConfig copies, upvotes, and views on AllMCPs.

How Open Compute (computer Use) works

What open-compute (computer-use) does

The open-compute (computer-use) MCP server gives an MCP client tools for observing a desktop, identifying application windows, inspecting Windows UI Automation elements, and performing controlled input. The client remains responsible for reasoning: it can capture an image, inspect the returned metadata or UIA tree, and then choose an action.

Its interaction model is intended for local computer use. The launcher communicates over MCP stdio and starts the open-compute engine through the local runtime. The engine connects to the desktop for screen capture, UI Automation, mouse and keyboard input, overlays, and voice-related operations.

How it works

A typical workflow begins with list_windows, capture, or tree. Window listings provide stable process and window identifiers, titles, rectangles, centers, and issued tokens. Captures return an observation identifier, which is needed when coordinate-based actions are requested. UIA inspection provides elements and a one-shot observation identifier for their coordinates.

Actions are safety-gated. do requires the relevant observation identifier and issued window descriptor or token for coordinate operations. click_name uses exact-first matching and reports ambiguity, scores, and alternatives. invoke activates a UIA element without clicking it. Safety behavior can be configured with OC_SAFETY_MODE, while OC_DENY can provide a hard action blacklist.

The remaining tools cover supporting workflows. watch_dir reports filesystem changes, push_status reads feed-manager status, and rec_replay replays a .clirec macro when the optional clirec package is available. Signal tools display or hide a pre-action countdown and overlay, report its state, or request an abort reason from a human. chat sends a human message about screen content, optionally with a screenshot. talk records a push-to-talk voice note as a WAV path for model-side speech processing.

Setup and configuration

Install and run the npm launcher in a local MCP client. The repository specifies Node.js 18 or newer and documents an npx open-compute-mcp launch path. The launcher starts the Python engine through uvx; OPEN_COMPUTE_MCP_CMD can be used by the launcher’s command-resolution configuration.

The MCP client and launcher must run on the host whose desktop is being controlled. An interactive Windows desktop session is required for real capture and input. Configuration also includes OC_SAFETY_MODE, with documented modes including confirmation, read-only, and allow-all, plus OC_DENY for prohibited actions.

Limitations and notes

The open-compute (computer-use) MCP server is local stdio software, not a hosted service. A deployment flow that cannot access the client host’s desktop cannot provide real screen capture or input. Coordinate actions depend on fresh observation data and issued window credentials, so callers should not treat coordinates as context-free.

UIA-specific targeting is intended for Windows desktop applications. rec_replay has an extra dependency, and voice input requires the push-to-talk interaction exposed by talk. Read-only tools such as screen geometry, window listing, and feed-manager status do not replace the safety requirements for actions.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Open Compute (computer Use)

open-compute (computer-use) is an MCP server that connects an MCP client to an interactive desktop, primarily for Windows capture and UI Automation. Its main tools include `capture`, `tree`, `list_windows`, `do`, `click_name`, and `invoke`, with additional tools for signals, file watching, status, replay, chat, and voice notes.

Install and launch the npm package locally with `npx open-compute-mcp` from an MCP client configuration. Node.js 18 or newer is specified, and the launcher starts the Python engine through `uvx`. The client must run on the desktop host that the tools will observe or control.

Yes, Cursor is named as an MCP client in the repository architecture. Configure the local stdio launcher in Cursor and run it on the host with the interactive desktop session. Hosted deployment is not appropriate because capture and input require access to that local desktop.

Actions are safety-gated through the engine and can be configured with `OC_SAFETY_MODE`. Coordinate-based calls require an observation ID plus an issued window descriptor or token, while `click_name` uses exact-first matching and reports ambiguity. `OC_DENY` can define a hard action blacklist.

No, the documented launcher is local stdio and real capture or input requires an interactive Windows desktop on the MCP client’s own host. A hosted deployment flow cannot access that desktop, so use a local MCP client instead.

Related MCP servers

All in OS Automation All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewOpen Compute (computer Use) AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/open-compute-computer-use?style=directory)](https://allmcps.com/mcp/open-compute-computer-use)
HTML Embed
<a href="https://allmcps.com/mcp/open-compute-computer-use"><img src="https://allmcps.com/api/badge/open-compute-computer-use?style=directory" alt="Open Compute (computer Use) on AllMCPs" /></a>