What makes this different from pixel‑based computer use tools?
It reads desktop‑rendered elements instead of pixels, making it faster and more reliable than vision‑based models, and runs natively on macOS without virtual machines.
@m13v
Basic information
No overview available yet
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"MCP-server-client-computer-use-ai-sdk": {
"command": "npx",
"args": [
"tsx",
"main.ts"
]
}
}
}Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
Fetch the live tool list by running this server in a temporary sandbox using the button above.
It is an open‑source MCP server that controls a computer natively on macOS, using underlying desktop‑rendered elements instead of pixel‑based vision. It is designed as an alternative to proprietary computer‑use agents like OpenAI’s Operator and Claude’s Computer Use, giving developers full control without virtual machines or guardrails.
Clone the repository, install Rust and Node.js, then run the backend server with cargo run --bin server in the mcp-server-os-level directory. Choose between a CLI interface (set the ANTHROPIC_API_KEY environment variable and run npx tsx main.ts in mcp-client-cli-interface) or a web‑app interface (set the key in a .env file and run npm run dev in mcp-client-nextjs).
It reads desktop‑rendered elements instead of pixels, making it faster and more reliable than vision‑based models, and runs natively on macOS without virtual machines.
You need Rust (install via rustup), Node.js and npm, and an Anthropic API key set as an environment variable.
The README only describes native macOS support; no other platforms are mentioned.
Set the `ANTHROPIC_API_KEY` environment variable (or place it in a `.env` file for the web app) with your Anthropic API key.
No guardrails — the tool is designed for full control of your computer, and you can use it with any app or website however you want.

One case file per task — shared by every session, agent and person that touches it. Self-hosted MCP server and live board. MIT.
Set up and manage the Eddyter AI rich text editor by talking to your AI agent — for Claude, Cursor, Codex and other MCP tools. A Model Context Protocol server that lets AI tools add Eddyter to a React, Next.js, Vite or R

Verified distributed compute for humans, applications, and agents. Submit Blender, Octane, and video workloads, route them across qualified GPUs, track asynchronous execution, download finished artifacts, and verify FARP

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.
Query your team's drift, vulnerability, and migration data from any AI assistant. Vibgrate MCP connects Cursor, Claude, ChatGPT, Windsurf, or VS Code to Vibgrate Cloud: 51 tools for DriftScores, CVEs and EOL runtimes, up
Comments