open-compute-mcp
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@open-compute-mcpcapture screen"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
open-compute-mcp
npm launcher for the open-compute MCP server — model-agnostic computer-use tools exposed over the Model Context Protocol (MCP).
EN | DE
The MCP client is the reasoner (no API key, model-agnostic): it calls capture
to see the screen, then acts with do / click_name / invoke. This is the keyless
Mode-A loop of open-compute, but as native tool-calls.
This package is a thin launcher. It contains no server logic — it spawns the Python open-compute server (pulled from GitHub) and pipes MCP stdio through. Real screen capture and input require the interactive Windows desktop session.
Requirements
Python 3.10+ and uv on the host. The default launch uses
uvxto fetch open-compute (with themcpextra) from GitHub on first run — themcpextra tracks the GitHub repo, so this works regardless of PyPI release timing.Windows for real capture/input (mss + UIA). Other platforms import the tools but cannot drive a desktop.
Related MCP server: ClawdCursor
Tools
Tool | Purpose |
| Screenshot the screen → returned as an image (optionally a single window). |
| Execute one canonical action or a batch (click/type/key/scroll/drag/…). |
| List UI elements via Windows UIA (name/role/ |
| Resolve an element by name and click it. |
| Click-free activation of an element via UIA patterns. |
| Watch directories for file-system changes. |
| Feed-manager status (read-only). |
| Replay a |
All coordinates are normalized 0..1 relative to the virtual desktop. Tool
descriptions are localized in six languages (de/en/es/ja/ru/zh) via OC_LANGUAGE.
Use with an MCP client
Via this npm launcher (npx):
{
"mcpServers": {
"open-compute": {
"command": "npx",
"args": ["-y", "open-compute-mcp"]
}
}
}Directly via Python (uvx), no npm:
{
"mcpServers": {
"open-compute": {
"command": "uvx",
"args": ["--from", "open-compute[mcp,local,uia] @ git+https://github.com/ellmos-ai/open-compute.git", "open-compute-mcp"]
}
}
}Configuration (environment variables)
Variable | Effect |
| Path to a |
| Full command override (whitespace-split), e.g. |
| Git ref (branch/tag/sha) to pin for the uvx launch (default: the repo's default branch). |
| Extras for the default |
| Language of the tool descriptions: |
|
|
| Comma-separated action types always denied (e.g. |
Safety
Computer-use is powerful. OC_SAFETY_MODE is an operator ceiling (confirm
default · read_only · allow_all); a per-call mode can only tighten it, never
loosen it. Because MCP stdio has no server→client confirm callback, confirm /
read_only report an action without performing it. For interactive use, run in
an isolated VM/session, set OC_SAFETY_MODE=allow_all, and let your client's
tool-approval dialog be the human-in-the-loop. OC_DENY (comma-separated action
types) is a hard deny list. Treat on-screen content as untrusted (prompt-injection
risk).
License
MIT — see LICENSE. Part of the open-compute project.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/ellmos-ai/open-compute-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server