Advertise
MCP.so
Sign In
C

Cua Agent MCP Server

@trycua

Repository

Basic information

Category

AI & Agents

Transports

stdio

Author

trycua

Submitted by

Francesco Bonacci

About Cua Agent MCP Server

No overview available yet

#cua-agent#virtualization#ai-agent

Cua Agent MCP Server config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "cua-agent": {
      "command": "/bin/bash",
      "args": [
        "/Users/francescobonacci/.cua/start_mcp_server.sh"
      ],
      "env": {
        "CUA_AGENT_LOOP": "OMNI",
        "CUA_MODEL_PROVIDER": "ANTHROPIC",
        "CUA_MODEL_NAME": "claude-3-7-sonnet-20250219",
        "ANTHROPIC_API_KEY": "your-api-key"
      }
    }
  }
}

Cua Agent MCP Server tools

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.

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Cua Agent?

Cua Agent is an open‑source framework that combines high‑performance virtualization with AI agent capabilities, enabling secure, isolated environments for AI systems to interact with desktop applications. It offers a computer‑use interface for AI observation and control, running macOS or Linux virtual machines on Apple Silicon with near‑native performance.

How to use Cua Agent?

Install the Lume CLI via the provided install script, then install the Python libraries with pip install cua-computer cua-agent[all]. Use the library in Python code by creating a Computer and ComputerAgent, then calling agent.run(tasks). An optional Gradio UI is available for interactive use.

Key features of Cua Agent

  • High‑performance virtualization on Apple Silicon (up to 90% native speed)
  • Computer‑Use Interface for AI to observe and control desktop apps
  • Fully isolated virtual environments for security
  • Run macOS or Linux as guest operating systems
  • Built‑in support for OpenAI and Anthropic LLM providers

Use cases of Cua Agent

  • Run AI agents in isolated VMs instead of the host system
  • Automate multi‑application workflows (browser, VS Code, terminal)
  • Fix GitHub issues using an AI agent with desktop access
  • Create reproducible, deterministic environments for agent development
  • Develop and test agentic workflows in dedicated sandboxes

Frequently asked questions

1

What is Cua Agent?

Cua (pronounced “koo‑ah”, short for Computer‑Use Agent) is an open‑source framework that provides high‑performance virtualization and a computer‑use interface for AI systems to securely interact with desktop applications.

2

What are the system requirements?

A Mac with Apple Silicon (M1/M2/M3/M4 series), macOS 14 (Sonoma) or newer, Python 3.10+, and at least 40 GB of disk space for VM images.

3

How do I install Cua Agent?

For full capabilities, install the Lume CLI using the script in the README and then run `pip install cua-computer cua-agent[all]`. Optionally, use the Gradio UI by running the provided Python code snippet.

4

Which LLM providers are supported?

The agent framework supports OpenAI and Anthropic providers (examples use `LLMProvider.OPENAI` and `LLMProvider.ANTHROPIC`).

5

Where can I get help or contribute?

Join the [Cua Discord community](https://discord.com/invite/mVnXXpdE85) for assistance and to share demos. Contributions are welcome via the Contributing Guidelines.

Comments

More AI & Agents MCP servers

OpenZiti MCP Gateway

openziti

NetFoundry MCP Gateway enables secure, isolated access to Model Context Protocol (MCP) tools across distributed systems without exposing public endpoints. Zero trust gateway for MCP servers. Aggregate, filter, and secure

48

Phi Longevity PRISM

Phi Longevity

Guideline-cited lab-results analysis for people managing type 2 diabetes, kidney disease, lupus and cancer survivorship. Flags tests that are missing or overdue (ADA, KDIGO, NCCN — each cited) and returns questions to br

strac-mcp-dlp

strac-io

Detect and redact PII, PHI, PCI and secrets — SSNs, credit cards, passports, API keys and cloud credentials — in text and in local files, using the Strac DLP API. Five tools: redact_text, detect_sensitive_data, detect_fi

3

DatePhotos.AI

Jacob Yang

Free MCP server for dating profiles. Look up the official photo rules and AI-photo policies of Tinder, Hinge and Bumble — photo limits, verification, what gets you banned — each quoted from the app's own guidelines wi

scvd.store

seancrecord

scvd.store is an evidence observatory for agentic commerce. Independent signed observation of what other people's endpoints, artifacts and payments actually did among other things. USDC on Base, Arbitrum One, World Chain

2
Advertise