# Browsermcp Ext

> A Gemini CLI extension that integrates BrowserMCP, giving your AI agent the ability to navigate and interact with web pages directly in your browser.

- **Type:** MCP server
- **Install:** `agentstack add mcp-derailed-dash-browsermcp-ext`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [derailed-dash](https://agentstack.voostack.com/s/derailed-dash)
- **Installs:** 0
- **Category:** [Web & Browser](https://agentstack.voostack.com/c/web-and-browser)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [derailed-dash](https://github.com/derailed-dash)
- **Source:** https://github.com/derailed-dash/browsermcp-ext

## Install

```sh
agentstack add mcp-derailed-dash-browsermcp-ext
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# BrowserMCP Extension for Gemini CLI

This is a Gemini CLI extension that integrates [BrowserMCP](https://browsermcp.io/), providing your AI agent with the ability to view, navigate, and interact with web pages directly in your browser.

## What This Extension Does

By giving the agent "eyes" and "hands", this extension enables the Gemini CLI to:

* Navigate to specific URLs
* Inspect and read the DOM
* Click buttons and links
* Type text into form fields
* Read browser console logs

This makes the extension incredibly useful for automating UI testing, extracting data from complex web apps, or having the agent perform repetitive web-based workflows on your behalf.

## Prerequisites & How It Works

This extension is a lightweight wrapper that connects the Gemini CLI to the BrowserMCP server.

For this to work, you **must** have the BrowserMCP Chrome extension installed in your active browser session:

1. Install the [BrowserMCP Chrome Extension](https://docs.browsermcp.io/setup-extension). (It takes about 30 seconds.)
2. Navigate to the web page you want the agent to interact with.
3. Click the BrowserMCP extension icon and click **"Connect"**.

*Note: The agent cannot spawn new browser sessions or operate in "headless" mode. It relies entirely on the active, connected tab.*

## Installation

To install this extension into your Gemini CLI, run:

```bash
gemini extensions install https://github.com/derailed-dash/browsermcp-ext
```

Once installed, simply start the Gemini CLI and run `/mcp` to verify the `browsermcp` tools are loaded. 

## Disclaimer

I am the author of this specific Gemini CLI extension wrapper. However, I am **not** the author of the BrowserMCP server or the underlying BrowserMCP project it relies upon. Full credit goes to the creators of [BrowserMCP](https://browsermcp.io/).

## Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [derailed-dash](https://github.com/derailed-dash)
- **Source:** [derailed-dash/browsermcp-ext](https://github.com/derailed-dash/browsermcp-ext)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/mcp-derailed-dash-browsermcp-ext
- Seller: https://agentstack.voostack.com/s/derailed-dash
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.