# GDD - AI-Controlled Browser Farm

Spins up isolated Chromium instances to test web apps across 22 simulated devices. Useful when an AI agent needs parallel browser automation.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 12
- Author: Cap-of-tea
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/Cap-of-tea/GDD
- Updated: 2026-10-14
- HTML: https://affpapa.org/en/nevibecode/mcp/gdd-ai-controlled-browser-farm

## What it does

The project orchestrates independent Chromium processes on your local machine. Every instance keeps its own cookies, storage cache, simulated geolocation, and network profiles. AI assistants interact with these environments through thirty-nine built-in MCP tools over HTTP on port `9700`.

Two distribution modes exist. The headless Server binary handles pure automation, while the Desktop application adds a visual monitor on port `9800` where you can click into any running viewport.

Without an active Docker daemon or the local binary listening on port `9700`, the assistant cannot interact with any web page.

## Requirements
- Needs: Docker

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add gdd-ai-controlled-browser-farm -- docker run -i --rm ghcr.io/cap-of-tea/gdd:1.5.0
```

### Codex

**Platform CLI**
```bash
codex mcp add gdd-ai-controlled-browser-farm -- docker run -i --rm ghcr.io/cap-of-tea/gdd:1.5.0
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "gdd-ai-controlled-browser-farm": {
            "command": "docker",
            "args": [
                "run",
                "-i",
                "--rm",
                "ghcr.io/cap-of-tea/gdd:1.5.0"
            ]
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "gdd-ai-controlled-browser-farm": {
            "type": "stdio",
            "command": "docker",
            "args": [
                "run",
                "-i",
                "--rm",
                "ghcr.io/cap-of-tea/gdd:1.5.0"
            ]
        }
    }
}
```

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add gdd-ai-controlled-browser-farm docker run -i --rm ghcr.io/cap-of-tea/gdd:1.5.0
```

## When to use it
- Run parallel user flows across desktop and mobile screen profiles
- Capture screenshots across multiple simulated screens simultaneously
- Inspect browser console logs for client-side JavaScript errors
- Isolate session storage and cookies across concurrent test agents

## Example prompts
- Launch three iPhone 15 Pro sessions and one desktop browser, open `https://myapp.com`, and test the sign-in flow.
- Take a screenshot of player 1 and check the browser console for errors.
- Click `#signup-btn` on the mobile instance and verify the response.

## Similar
- [Front-End Checklist](https://affpapa.org/en/nevibecode/mcp/front-end-checklist.md) — MCP server · ★ 74569
- [Keploy](https://affpapa.org/en/nevibecode/mcp/keploy.md) — MCP server · ★ 18779
- [SeleniumBase MCP](https://affpapa.org/en/nevibecode/mcp/seleniumbase-mcp.md) — MCP server · ★ 13181
- [mobile-mcp](https://affpapa.org/en/nevibecode/mcp/mobile-mcp.md) — MCP server · ★ 10025
- [e2e](https://affpapa.org/en/nevibecode/mcp/e2e.md) — MCP server · ★ 11962
- [mockserver](https://affpapa.org/en/nevibecode/mcp/mockserver.md) — MCP server · ★ 4999

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md