Best for
- MCP clients that need browser automation or browser access
- Local or remote environments using Chrome, Edge, or Firefox
- Workflows that connect to a browser through CDP or WebSocket
The source lists this server as active at version 1.0.0 under the Apache-2.0 license. It is focused on browser-use access; the source does not list specific MCP tools, client compatibility, or authentication mechanisms.
Strongest · Maintenance 12/20 Weakest · Reliability 8/20
Static review · not runListed 2026-07-29
Read the FMRS scoring method →Runtime:Node.js 22+
DISPLAY
DISPLAY environment variable for browser rendering on Linux; usually set automatically in a desktop session (e.g. :0), no manual setup needed.
Use the npm package @agent-infra/mcp-server-browser. For stdio, run the package directly; for SSE and Streamable HTTP, use npx and set the port parameter, which defaults to 8089. Configure the browser channel, CDP or WebSocket endpoint, browser executable path, output directory, proxy settings, and vision mode as needed.
Run npx @agent-infra/mcp-server-browser and check that its browser tools appear in your MCP client's tool list; issue a request to open a page and confirm a snapshot is returned to prove the connection works.
Once connected, you can ask your AI assistant things like:
This is the browser MCP server in ByteDance's UI-TARS-desktop repository. It provides browser-use access and can run over stdio, SSE, or Streamable HTTP. It supports Chrome, Edge, and Firefox, with optional CDP or WebSocket connections. Aria snapshots are used by default, with an optional screenshot-based vision mode.
Microsoft's official Playwright-based browser automation MCP server
The Chrome team's official MCP server exposing Chrome DevTools capabilities to AI assistants
Control the FSB browser extension from any MCP client for DOM-first AI browser automation.
Playwright-based browser automation MCP server
Source revision 3ac2cb8d946d Data synced 2026-10-19 Read the FMRS scoring method