Skip to main content
Glama
Sign In
deenesjakoruzh
browser-use-native-windows

browser-use-native-windows

by mlnima
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

browser-use-native-windows

Windows-only MCP server for controlling a real Chromium browser with native screenshots, Windows accessibility, and Windows mouse/keyboard input.

It does not use CDP, Chrome DevTools, Playwright, Puppeteer, browser extensions, DOM selectors, DOM snapshots, or page JavaScript evaluation.

Every observation re-detects the browser window, monitor, physical resolution, monitor scaling, and window position. Window and monitor geometry is verified again before native mouse input; resize or scaling changes invalidate the observation so the client must observe the new screenshot before acting.

Requirements

  • Windows

  • Node.js 20+

  • Chromium-based browser: Edge, Chrome, Brave, Chromium, Vivaldi, Opera, Yandex

Related MCP server: hermes-computer-use

Install

From source:

npm install
npm run build

Optional global install from this package root:

npm install -g .

Configuration

The MCP reads system environment variables first. If a .env file exists next to this README, it is loaded as a fallback. The MCP starts normally when .env is missing.

Create .env from .env.example when you want fixed HTTP or browser settings:

BROWSER_USE_NATIVE_WINDOWS_SSE_HOST= "0.0.0.0"
BROWSER_USE_NATIVE_WINDOWS_SSE_PORT= "7331"
BROWSER_USE_NATIVE_WINDOWS_SSE_AUTH= "change.me"
BROWSER_USE_NATIVE_WINDOWS_BROWSER_EXECUTABLE_PATH= "C:\Program Files (x86)\Microsoft\Edge\Application\msedge.exe"
BROWSER_USE_NATIVE_WINDOWS_BROWSER_USER_DATA_DIR= "C:\Users\YOUR_USER\AppData\Local\Microsoft\Edge\User Data"
BROWSER_USE_NATIVE_WINDOWS_PAGE_LOAD_TIMEOUT_MS= "10000"

Change BROWSER_USE_NATIVE_WINDOWS_SSE_AUTH before exposing HTTP outside your machine.

Run

Stdio transport:

npm run start:stdio

Streamable HTTP at /mcp:

npm run start:mcp

Legacy HTTP+SSE at /sse with POST messages at /messages:

npm run start:sse

Both network transports:

npm run start:all

Global install:

browser-use-native-windows
browser-use-native-windows --transport mcp
browser-use-native-windows --transport sse
browser-use-native-windows --transport all

The default network host is 0.0.0.0, so /mcp and /sse accept connections through localhost and this computer's LAN address. Bearer authentication is required on every network endpoint.

MCP Client

Stdio:

{
  "mcpServers": {
    "browser-use-native-windows": {
      "transport": "stdio",
      "command": "node",
      "args": ["<package-root>\\dist\\index.js"],
      "cwd": "<package-root>"
    }
  }
}

Codex Streamable HTTP:

[mcp_servers.browser_use_native_windows]
url = "http://<host>:7331/mcp"
http_headers = { "Authorization" = "Bearer change.me" }

SSE compatibility:

{
  "mcpServers": {
    "browser-use-native-windows": {
      "transport": "sse",
      "url": "http://<host>:7331/sse",
      "headers": {
        "Authorization": "Bearer change.me"
      }
    }
  }
}

Tools

  • browser_observe: launch or adopt the browser and return a native observation.

  • browser_act: run one mouse or keyboard action against a fresh observation token.

  • browser_status: return transport, driver, browser, window, focus, monitor, DPI, and observation state.

  • browser_stop: release held input state and optionally close the tracked browser.

Force Stop

Default global hotkey:

Control+F12

The watchdog releases held keys and mouse buttons, then stops the MCP process.

Available Tools

4 tools
browser_actAct In BrowserB

Run one native mouse or keyboard action against a matching fresh browser_observe token.

Inputs: observationToken (required), action (required).

browser_observeObserve BrowserC

Launch or adopt a Windows browser, optionally handle a target URL through native UI, and return a browser-window or browser-owned file-dialog observation.

Inputs: targetUrl, inlineImage.

browser_statusBrowser StatusB

Return MCP transport, native input driver, browser process, HWND, monitor, DPI, focus, and observation state.

Inputs: no named parameters.

browser_stopStop Browser ControlA

Release held native input state and optionally close the tracked browser only when the user task asks for it.

Inputs: closeBrowser.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    A3.7/5.0

    Scored across 4 tools

    Disambiguation5/5

    Each tool has a clearly distinct role: browser_observe for launching/adopting and getting observation, browser_act for performing actions, browser_status for checking state, and browser_stop for cleanup. No overlap in purpose.

    Naming Consistency5/5

    All tool names follow a consistent 'browser_' prefix followed by a single verb (observe, act, status, stop), all lowercase, with no mixing of conventions.

    Tool Count5/5

    Four tools is exactly right for this domain: observe, act, status, and stop cover the essential lifecycle without unnecessary extras. Not too few, not too many.

    Completeness5/5

    The tool set covers the full lifecycle of native browser interaction: initialization (browser_observe), action (browser_act), monitoring (browser_status), and termination (browser_stop). No obvious gaps.

    Resources

    Related MCP Connectors

    Related MCP Servers

    • computer-use-windows

      wimi321
      A
      license
      Not graded
      quality
      D
      maintenance
      A standalone MCP server for Windows desktop control, enabling screenshots, mouse and keyboard input, app launch, window/display management, and clipboard access via natural language.
      3
      MIT

    • windows-computer-use-mcp

      Einosensokami
      A
      license
      Not graded
      quality
      D
      maintenance
      A local, dependency-free MCP server that gives AI agents controlled access to the active Windows desktop, enabling automated interaction with applications through screenshots, clicks, typing, and window management.
      151 npm
      1
      MIT

    • pc-control-agent

      anbu-20012005
      F
      license
      Not graded
      quality
      D
      maintenance
      This MCP server enables an AI agent to control a Windows PC through human-like interactions such as screen capture, OCR, mouse, keyboard, and navigation, without using shortcut APIs.
      -