Skip to content

the-agents-work/taw-computer

Repository files navigation

taw-computer

Your AI writes code. This lets it use a computer.

Stars  MIT License  Issues  Get Started

Quick Start · Demo · Hosted Version · Contributing · Report Bug


What if your AI could do everything you do on a computer?

Not just write code — but open a browser, click buttons, fill forms, run servers, test in real browsers, install anything, and see the screen?

taw-computer is an open-source MCP server that gives AI agents a full Ubuntu desktop inside Docker. Your AI connects, gets a real computer, and works like a human would.

No internal LLM. No chat UI. Your AI is the brain. This is the body.


Demo

📹 Demo coming soon — star this repo to get notified!


Why taw-computer?

Other tools let AI write code. taw-computer lets AI use a computer.

ChatGPT / Claude Cursor / Copilot Lovable / Bolt taw-computer
Write code
Run shell commands Limited Sandboxed Full Ubuntu
Browse the web Real Chromium
See & click the screen Desktop + VNC
Install any software apt/npm/pip
Test in real browser Preview only Playwright + CDP
Persist across sessions Snapshots
Self-hostable 100% yours

Quick start

Get running in under 5 minutes:

# 1. Clone & build
git clone https://github.com/the-agents-work/taw-computer.git
cd taw-computer
docker build -f images/Dockerfile.taw -t taw-computer-base .

# 2. Install & start
npm install && npm start

Then add to your AI client:

Claude Code (~/.claude/mcp.json)
{
  "mcpServers": {
    "taw-computer": {
      "command": "npx",
      "args": ["tsx", "/path/to/taw-computer/mcp/index.ts"]
    }
  }
}
Cursor

Add to Cursor MCP settings (Settings → MCP Servers) — same JSON format as above.

Claude Desktop

Add to claude_desktop_config.json — same JSON format as above.

Any MCP client

taw-computer speaks standard MCP over stdio. Any client that supports MCP can connect.

Remote server (SSH) — run on a beefy machine, use from your laptop

Got a powerful server / Mac Mini / VPS? Run taw-computer there and connect from anywhere:

{
  "mcpServers": {
    "taw-computer": {
      "command": "ssh",
      "args": ["user@your-server", "cd /path/to/taw-computer && npx tsx mcp/index.ts"]
    }
  }
}
Your laptop (Claude Code)
    ↕ SSH (stdin/stdout piped over network)
Remote server (taw-computer + Docker)
    ↕ Docker
Ubuntu sandbox

Setup:

  1. On the server: install Docker, clone repo, build image, npm install
  2. On the server: enable SSH (sudo systemctl enable ssh)
  3. On your laptop: ssh-copy-id user@your-server (passwordless login)
  4. Add the MCP config above — done!

Watch via VNC: open http://your-server:6080 in your browser.

That's it. Now tell your AI: "Create a VM and build me a website" — and watch it work.


What can it do?

🖥️ "Build me a landing page"

AI creates a VM → scaffolds Next.js → writes components → starts dev server → opens browser to check → iterates until it looks right

🌐 "Go to Amazon and find the best laptop under $1000"

AI opens Chromium → navigates to Amazon → searches → scrolls → extracts prices → compares → reports back

🧪 "Run E2E tests on my deployed app"

AI launches Playwright → navigates to your URL → fills forms → clicks buttons → asserts results → reports failures

🔧 "Set up a PostgreSQL database with sample data"

AI runs apt install postgresql → creates database → writes seed script → runs it → verifies with queries

📸 "What does my app look like on mobile?"

AI takes desktop screenshot → resizes viewport → screenshots again → compares → suggests CSS fixes


How it works

┌─────────────────────────────────────────────────────┐
│  Your AI Client                                     │
│  Claude Code · Cursor · Claude Desktop · any MCP    │
└───────────────────────┬─────────────────────────────┘
                        │ MCP protocol (stdio)
┌───────────────────────▼─────────────────────────────┐
│  taw-computer MCP server          30+ tools         │
│  vm · shell · files · browser · desktop · search    │
└───────────────────────┬─────────────────────────────┘
                        │ Docker API
┌───────────────────────▼─────────────────────────────┐
│  Ubuntu 22.04 Sandbox              isolated container│
│                                                      │
│   bash    Chromium + CDP    xfce4 Desktop + VNC     │
│     git npm pip curl          Playwright              │
│       python node              xdotool scrot          │
│                                                      │
│   /workspace ← your project files live here          │
└──────────────────────────────────────────────────────┘

30+ tools

VM Management — create, destroy, snapshot, resume
Tool What it does
vm_create Spin up a new sandbox. Returns VNC URL to watch live
vm_destroy Destroy (auto-saves snapshot for later)
vm_reset Destroy + delete snapshot (fresh start)
vm_restart Restart container, keep all files
vm_status CPU, RAM, disk, uptime, top processes
vm_list List running sandboxes
vm_rename Rename a VM
snapshot_list List saved snapshots
snapshot_delete Delete a snapshot
Shell & Files — full Ubuntu command line + file ops
Tool What it does
exec Run any command: git, npm, pip, curl, docker, anything
fs_read Read a file
fs_write Write a file (creates parent dirs)
fs_edit Find-and-replace in a file
fs_list ls / recursive find
fs_search grep for patterns
code_search ripgrep with regex, file types, context
file_upload Upload file into VM (base64, max 50MB)
Browser (CDP/Playwright) — real browser, not a simulator
Tool What it does
browser_navigate Go to URL, wait for load
browser_snapshot Screenshot + numbered overlays on every clickable element
browser_click_ref Click element #N from snapshot
browser_type_ref Type into element #N
browser_extract Read page text (CSS selector or full page)
browser_eval Run JavaScript in page
browser_wait_for Wait for selector / text / network idle
browser_console_logs Read console.log, console.error, etc.
browser_network_errors Catch 404s, CORS errors, failed requests
browser_run_test Run a Playwright test script
browser_open Open Chrome via desktop (fallback)
browser_close Kill Chrome
web_search Google search → top 8 results
Desktop — see and control the GUI
Tool What it does
desktop_screenshot JPEG screenshot of the whole desktop
desktop_click Click at (x, y)
desktop_type Type text into focused window
desktop_key Key combos: ctrl+c, alt+tab, Return, etc.
desktop_scroll Scroll up/down
desktop_drag Drag from A to B

Set-of-Mark: how browser automation actually works

Most "computer use" tools guess pixel coordinates. We use Set-of-Mark prompting — the AI sees numbered badges on every interactive element:

Step 1: browser_snapshot
        → AI sees screenshot with [1] Login  [2] Search  [3] Cart  ...

Step 2: browser_click_ref(ref=2)
        → clicks the Search box precisely

Step 3: browser_type_ref(ref=2, text="laptop", submit=true)  
        → types and presses Enter

Step 4: browser_snapshot
        → sees new page with results [4] [5] [6] ...

No coordinate guessing. No CSS selector fragility. The AI sees what it's clicking.


VNC — watch your AI work in real time

Every sandbox comes with a noVNC web viewer. Open the URL in your browser and watch:

  • 🖱️ AI navigating websites and clicking buttons
  • ⌨️ AI writing code in the terminal
  • 🏗️ AI building and testing applications
  • 🐛 AI debugging by inspecting the screen

Perfect for demos, debugging, and building trust in AI agents.


What's inside each sandbox

Included
OS Ubuntu 22.04
Desktop xfce4 + Xvfb + x11vnc + noVNC
Browser Playwright Chromium (native arm64 + amd64)
Languages Node.js 20, Python 3, build-essential
CLI git, curl, wget, jq, ripgrep, tree, nano, vim
DB clients PostgreSQL, MariaDB, Redis
Dev tools GitHub CLI, yq, httpie
Automation xdotool, scrot, imagemagick, xclip

Configuration

Variable Default Description
MAX_SANDBOXES 3 Max concurrent VMs
SANDBOX_TYPE auto auto / docker / firecracker
DOCKER_IMAGE taw-computer-base Base image
DOCKER_MEMORY_MB 4096 RAM per container
DOCKER_CPUS 2 CPUs per container
DESKTOP_RESOLUTION 1280x720 Screen resolution

Requirements

Minimum
Docker Docker Desktop or Docker Engine
Node.js 20+
RAM ~4GB per sandbox
Disk ~5GB for base image

Project structure

taw-computer/
├── mcp/
│   ├── index.ts            # MCP server — stdio, 30+ tool handlers
│   └── browser.ts          # Playwright CDP + Set-of-Mark engine
├── sandbox/
│   ├── SandboxManager.ts   # Abstract interface
│   ├── DockerSandbox.ts    # Docker implementation
│   ├── FirecrackerSandbox.ts # Firecracker microVM (optional)
│   ├── NetworkManager.ts   # Network isolation
│   ├── config.ts           # Env-based config
│   └── index.ts            # Auto-detect backend
├── images/
│   └── Dockerfile.taw      # Ubuntu sandbox image
├── .github/
│   ├── workflows/ci.yml    # CI: typecheck + Docker build
│   └── ISSUE_TEMPLATE/     # Bug report + feature request
├── package.json
├── CONTRIBUTING.md
└── LICENSE (MIT)

Contributing

We'd love your help! See CONTRIBUTING.md.

Ideas for first contributions:

  • 🎨 Record a demo GIF for this README
  • 📝 Write a tutorial ("Build X with taw-computer")
  • 🔧 Add a new MCP tool (audio? clipboard? multi-tab?)
  • 🐳 Build a slimmer Docker image
  • 🧪 Add automated tests
  • 📦 Support Podman / containerd

Hosted version

Don't want to self-host? shipkit.cc — managed taw-computer with:

  • Chat UI (just type what you want)
  • Auth & team collaboration
  • One-click app sharing
  • No Docker setup needed

FAQ

How is this different from Lovable / Bolt / v0?

Those are closed-source, hosted-only products that generate code. taw-computer gives AI a real computer — it can run servers, browse the web, install anything, and interact with any desktop app. It's also open source and self-hostable.

How is this different from OpenInterpreter / Open Hands?

OpenInterpreter runs code on your local machine (risky). Open Hands uses its own LLM orchestration. taw-computer is just the computer — no built-in LLM, no opinions about orchestration. Your existing AI client (Claude Code, Cursor, etc.) is the brain. taw-computer is a pure MCP server.

Is it safe? Can the AI break my system?

Each sandbox is an isolated Docker container with its own filesystem, network, and process space. Nothing inside can touch your host system. Containers have memory/CPU/PID limits. When you're done, destroy the VM.

Can I use it with GPT-4 / Gemini / local models?

Yes — any AI client that supports MCP can connect. The server doesn't care which LLM is behind the client.

Does it work on Mac / Windows / Linux?

Yes. Anywhere Docker runs, taw-computer runs. The sandbox image supports both arm64 (Apple Silicon) and amd64 (Intel/AMD).


Star History

Star History

If taw-computer is useful to you, give it a ⭐ — it helps others find it.


Related

License

MIT — do whatever you want with it.

About

Give any AI a real computer. Open source MCP server with Ubuntu sandbox, browser automation, desktop control, and 30+ tools. Works with Claude Code, Cursor, Claude Desktop.

Topics

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors