# BrowserMCP/mcp

Browser MCP is a Model Context Provider (MCP) server that allows AI applications to control your browser

Repository: https://github.com/BrowserMCP/mcp
Canonical: https://ross.abutalabs.com/products/browsermcp-mcp
Homepage: https://browsermcp.io
Language: TypeScript
License: Apache-2.0
License Family: permissive
Topics: browser-automation, browser-extension, claude, cursor, mcp, mcp-server
Last push: 2025-04-24T21:49:44+00:00

## Health v2 (maintenance only)
Score: 28/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 18, release rhythm 35, longevity 37
- inputs: {"age_days": 524, "days_push": 496, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7019, forks 554 (observed 2026-08-28T04:09:53.426149+00:00)

## What it is
Browser MCP is an MCP server plus Chrome extension that lets AI applications like Claude, Cursor, VS Code, and Windsurf control the user's existing browser. It is adapted from Microsoft's Playwright MCP server but automates the user's real browser profile instead of launching new instances, keeping sessions logged in and avoiding basic bot detection.

## Use cases
- automate my browser with AI
- let Claude or Cursor fill out web forms for me
- run end-to-end UI tests from my AI code editor
- automate repetitive web tasks like data collection
- use my logged-in browser sessions with AI automation
- avoid CAPTCHAs and bot detection during browser automation

## When to choose
- you want AI apps to drive your real, logged-in browser profile
- you need local, private browser automation without remote servers
- you want to avoid basic bot detection and CAPTCHAs by using a real browser fingerprint
- you use Claude, Cursor, VS Code, or Windsurf and want browser control via MCP

## When to avoid
- you need headless automation or spawning fresh isolated browser instances (use Playwright MCP instead)
- you need a fully self-contained build - the repo depends on private monorepo code and cannot be built standalone
- you need large-scale scraping or parallel automation across many sessions

## Facets
- artifact type: service
- maturity: active
- function: mcp, web-scraping, workflow-automation, e2e-testing, browser-extension
- domain: developer-tools, web-development, browser-extensions
- platform: browser, browser-extension, cross-platform
- tags: mcp-server, browser-automation, chrome-extension, ai-agents, claude, cursor, playwright-mcp-fork, automation, nodejs

## Member repositories
- BrowserMCP/mcp (main) score 28

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:53.426149+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:40:38.024065+00:00, confidence not recorded.
  - readme: https://github.com/BrowserMCP/mcp (fetched 2026-08-28T04:09:53.426149+00:00, sha 3aff65195a33)
  - homepage: https://browsermcp.io (fetched 2026-08-29T08:36:51.441013+00:00, sha c7931a0ba953)
  - site_page: https://browsermcp.io/install (fetched 2026-08-29T08:36:51.452512+00:00, sha 606d651f186c)
  - site_page: https://docs.browsermcp.io/ (fetched 2026-08-29T08:36:51.454535+00:00, sha 845626513b46)
- Data as of 2026-08-30T08:39:29.467469+00:00.