Yetibrowser MCP

YetiBrowser MCP is a fully open - source implementation of browser automation MCP. It connects through a local MCP server and a browser extension, enabling AI assistants to control real browser tabs. All data is processed locally to ensure transparency and security.

Browser automation Developer tools #Browser automation #Local MCP #Open - source tool #AI assistant integration .TypeScript

rating : 2 points

downloads : 7.7K

update time : 2025-12-12

Open Site

What is YetiBrowser MCP?

YetiBrowser MCP is a locally - prioritized browser automation solution that allows AI assistants (such as Claude, Cursor, etc.) to control your real browser tabs through the MCP protocol. It consists of two parts: an MCP server running on your computer and a browser extension installed on Chrome or Firefox. AI assistants can navigate web pages, click buttons, fill out forms, take screenshots, etc. through this system, just like a real person operating, but all operations are completed on your local device, ensuring privacy and security.

How to use YetiBrowser MCP?

Using YetiBrowser MCP requires three steps: 1) Install the extension in your browser; 2) Configure the MCP server in your AI tool; 3) Send instructions through the AI assistant to control the browser. The extension will establish a connection with the local server. The AI assistant sends commands through the MCP protocol, and the server forwards these commands to the browser for execution. The entire process does not require cloud services, and the data is completely localized.

Use cases

YetiBrowser MCP is particularly suitable for the following scenarios: web page automation testing, data scraping and organization, automation of repetitive web page operations, web page content analysis and debugging, and assisting web development and QA testing. Whether developers are testing website functions or ordinary users need to automate certain web page tasks, they can easily achieve this through AI assistants.

Main features

Browser navigation control

AI can control the browser to access specified URLs, move forward and backward, refresh pages, and perform other basic navigation operations.

Page element interaction

Supports rich page interaction functions such as clicking buttons, entering text, selecting options, and dragging elements.

Page state capture

Can obtain page information such as page snapshots, screenshots, console logs, and form states to help AI understand the current page situation.

Intelligent waiting mechanism

Supports waiting for a specific element to appear or waiting for a specified time to ensure that subsequent operations are performed after the page has finished loading.

Form automation filling

Can fill multiple form fields at once, including text boxes, drop - down boxes, checkboxes, radio buttons, etc.

Dialog box handling

Can automatically handle browser dialog boxes such as alert, confirm, and prompt.

Custom JavaScript execution

Allows custom JavaScript code to be executed in the page context and obtain the results.

Page change comparison

Can compare the differences between two page snapshots and identify changes in DOM and ARIA attributes.

Advantages

Completely open - source and transparent: All code can be viewed, audited, and modified, with no hidden binary files.

Locally - prioritized: All browsing data is retained on your device and will not be sent to the cloud.

Privacy - secure: The extension only communicates with the local server to protect your browsing privacy.

Cross - browser support: Supports Chrome and Firefox browsers.

Developer - friendly: Provides rich debugging tools, such as console log capture and DOM difference comparison.

Production - ready: Includes complete packaging, publishing, and integration documentation.

Limitations

Limited Firefox support: Due to the incomplete support of Manifest V3 in Firefox, the functions of the Firefox version may be limited.

Requires manual configuration: You need to configure the MCP server connection in your AI tool.

Depends on local connection: The extension must be able to establish a WebSocket connection with the local server.

Requires AI tool support: Your AI assistant needs to support the MCP protocol to use it.

How to use

Install the browser extension

Install the corresponding extension according to your browser type. Chrome users can install it from the Chrome Web Store, while Firefox users need to wait for the full - version release.

Configure the MCP server

Configure the YetiBrowser MCP server in your AI tool. The configuration methods vary slightly for different tools. The following provides configuration examples for common tools.

Start the server and connect

Start the MCP server and ensure that the browser extension is connected to the correct port. The extension will automatically search for an available port by default, or you can set it manually.

Control the browser through the AI assistant

In the configured AI tool, you can now control the browser through natural language instructions. For example, ask the AI to navigate to a certain website or perform specific operations.

Usage examples

Automated web search

Let the AI assistant automatically open a search engine, enter a query term, and obtain search results.

Form auto - filling

Automatically fill out complex web forms to save manual input time.

Web content scraping and analysis

Scrape web data and perform analysis and organization.

Web testing and debugging

Automatically test web page functions and collect debugging information.

Frequently Asked Questions

Is YetiBrowser MCP secure? Will my browsing data be sent to the cloud?

Do I need programming knowledge to use it?

Which browsers are supported?

How to check the connection status?

What if there is a port conflict?

Which AI tools are supported?

Related resources

GitHub repository

The complete source code and documentation of YetiBrowser MCP.

Chrome Extension Store

The installation page for the Chrome browser extension.

MCP Protocol Documentation

The official specification documentation for the Model Context Protocol.

Workspace Command Documentation

Detailed instructions on development and workspace commands.

The privacy policy description of YetiBrowser MCP.

🚀 YetiBrowser MCP

YetiBrowser MCP is a fully open - source implementation of the Browser MCP workflow. It connects a Node - based MCP server with Chrome/Firefox extensions. This allows Model Context Protocol clients such as Codex/Claude Code, Cursor, Windsurf, MCP Inspector, or your own tools to automate a real browser tab. All the data stays on your machine and can be audited.

🚀 Quick Start

YetiBrowser MCP is designed to provide a seamless experience for automating browser tabs. To get started, you need to install the browser extension and the MCP server. The following sections will guide you through the installation process for different browsers and MCP - aware clients.

✨ Features

Why pick YetiBrowser MCP?

Transparent and hackable: There are no blob downloads. You can inspect, fork, and extend every component.
Local - first: The extension only communicates with a localhost MCP server. Browsing data never leaves your device.
Cross - browser: Shared logic powers both Chrome and Firefox packages. Note that the Firefox build is pending better Manifest V3 support, so the connection user experience may be limited until Mozilla releases full MV3 APIs.
Developer - focused tooling: It offers richer console capture, DOM diffing, page - state dumps, and full - page screenshots for debugging and QA.
Production - friendly: It comes with scripts and docs for packaging, publishing, and integrating with IDE workflows.

Repository layout

packages/shared: Shared TypeScript definitions for messages and tool schemas.
packages/server: The MCP server that bridges MCP clients to a running browser tab.
extensions/shared: Shared extension source (background/popup) and assets.
extensions/chrome / extensions/firefox: Per - browser packaging layers.
docs/: Workspace commands, publishing checklists, and feature notes.
scripts/: Helper utilities such as package - extensions.sh for release zips.

💻 Usage Examples

MCP Tools Available

browser_snapshot: Capture an accessibility - oriented snapshot of the current page.
browser_snapshot_diff: Compare the two most recent snapshots to highlight DOM/ARIA changes.
browser_navigate: Load a new URL in the connected tab and return an updated snapshot.
browser_go_back / browser_go_forward: Move through history while keeping MCP in sync.
browser_wait: Pause automation for a set number of seconds.
browser_wait_for: Block until a selector appears (optionally visible) before proceeding.
browser_press_key: Simulate a keyboard key press on the focused element.
browser_click: Click the element identified by a CSS selector.
browser_hover: Hover the pointer over the targeted element.
browser_drag: Drag an element onto a drop target for sortable/drag - and - drop UIs.
browser_type: Type text (optionally submitting with Enter) into an editable element.
browser_fill_form: Fill multiple inputs/selects/checkboxes/radios in a single call.
browser_select_option: Choose one or more options in a <select> element.
browser_screenshot: Capture a viewport or full - page screenshot via the DevTools protocol.
browser_get_console_logs: Return recent console output, including errors with stack traces.
browser_page_state: Dump forms, storage keys, and cookies for the connected page.
browser_connection_info: Report bridge WebSocket port, connection status, and extension version.
browser_evaluate: Run custom JavaScript inside the page and return JSON - serializable results.
browser_handle_dialog: Accept or dismiss alert/confirm/prompt dialogs with optional prompt text.

📦 Installation

MCP Browser Extension Installation

Chrome Browser

Install the latest Chrome Browser Extension here: https://chromewebstore.google.com/detail/hpedknkjljpopinabjjmljgckhlancni?utm_source=item - share - cb
Leave it on Automatic connection mode unless you are running multiple instances of YetiBrowser MCP Server

Firefox Browser (not currently available)

Due to the limited capabilities of Manifest v3 in Firefox stable release, the Firefox extension is on hold until full support is commonly available (outside of Nightly builds).

MCP Server Installation

Codex CLI

Edit your ~/.codex/config.toml and add the MCP entry:

[mcp_servers.yetibrowser - mcp]
command = "npx"
args = ["-y", "@yetidevworks/server"]

Restart codex CLI command; you should see yetibrowser - mcp listing under /mcp tools.
If you want to provide a specific port, use this format for the args entry: args = ["-y", "@yetidevworks/server", "--ws - port", "9010"]

Claude Code

Make sure the extension is installed and connected to a tab, then start the MCP server with npx -y @yetidevworks/server (or run the locally built CLI).
Add the server entry to ~/Library/Application Support/Claude/claude_desktop_config.json (see the example in ).
Restart claude so it picks up the new MCP server; you should see yetibrowser - mcp listed under the /mcp tools menu once the extension connects.

Other MCP - aware clients

Any MCP client can connect by spawning the CLI (npx -y @yetidevworks/server) or optionally provide a specific port, e.g., npx -y @yetidevworks/server --ws - port 9010.
The server exposes the standard MCP transport over stdio, so use whatever configuration mechanism your client supports to run the command above when a tab is connected.

MCP Inspector

For testing and debugging outside a coding agent.
Run npx @modelcontextprotocol/inspector npx -y @yetidevworks/server -- --ws - port 9010 to inspect the MCP server in conjunction with the YetiBrowser MCP browser extension.

Troubleshooting

The CLI walks ports 9010 - 9020 until it finds a free one, logging switched to when it advances. Pass --ws - port <port> if you want to pin a specific port instead.
The Browser extension popup mirrors that behaviour: leave it on “Automatic” to track the CLI’s port, or choose “Manual” and enter the port reported by browser_connection_info / the CLI log to override it.
Simply ask your AI tool about the connection port. For example: what is the yetibrowser mcp connection info?, then set the manual port to the reported port.

📚 Documentation

Workspace commands live in [docs/workspace - commands.md](docs/workspace - commands.md).
Publishing steps (npm + extension stores) are in .
Screenshot behaviour is documented in .
Generate distributable Chrome/Firefox zips with ./scripts/package - extensions.sh (outputs to artifacts/).
A repository - level privacy policy is available in .