MCP servers

Browser & Web Scraping MCP servers

Drive a real browser, take snapshots of pages and turn websites into clean text or structured data. 14 servers, 11 of them official, with the command to add each one, GitHub stars and the last push. Updated 2026-10-20.

Browser automation & web scraping

14

14 of 14 servers

  • Browser UseOfficial · Browser Use

    An agent that drives a real Chrome to finish tasks: fill forms, click through flows, extract data.

    ★ 119kpushed 2026-10-20MIT

    Local (stdio)
    claude mcp add browser-use -- uvx browser-use --cli-mcp
  • Chrome DevToolsOfficial · Google

    Control a live Chrome: performance traces, network requests, console and screenshots for debugging.

    ★ 54kpushed 2026-10-20Apache-2.0

    Local (stdio)
    claude mcp add chrome-devtools -- npx -y chrome-devtools-mcp@latest
  • PlaywrightOfficial · Microsoft

    Browser automation through the accessibility tree: click, type, navigate and snapshot pages, no screenshots needed.

    ★ 38kpushed 2026-10-19Apache-2.0Microsoft on DevHunt

    Local (stdio)
    claude mcp add playwright -- npx -y @playwright/mcp@latest
  • SkyvernOfficial · Skyvern

    AI browser automation in Skyvern's cloud: navigate, fill forms and pull data from any site.

    ★ 23kpushed 2026-10-20AGPL-3.0on DevHunt

    Remote
    claude mcp add --transport http skyvern https://api.skyvern.com/mcp/

    Headers: x-api-key (add with --header where the client doesn't sign in with OAuth)

  • ApifyOfficial · Apify

    Run any of thousands of Apify scrapers (Actors) for social media, maps, e-commerce and more.

    ★ 12kpushed 2026-10-20MIT

    Remote
    claude mcp add --transport http apify https://mcp.apify.com/

    Headers: Authorization (add with --header where the client doesn't sign in with OAuth)

  • FirecrawlOfficial · Firecrawl

    Scrape, crawl, map and search websites into clean Markdown or JSON.

    ★ 7.6kpushed 2026-10-20MITon DevHunt

    Remote + local
    claude mcp add --transport http firecrawl https://mcp.firecrawl.dev/v2/mcp

    Headers: Authorization (add with --header where the client doesn't sign in with OAuth)

    Or run it locally: npx -y firecrawl-mcp

  • Bright DataOfficial · Bright Data

    Web search and scraping that gets past blocks and CAPTCHAs, with a free monthly quota.

    ★ 2.7kpushed 2026-10-20MIT

    Local (stdio)
    claude mcp add --env API_TOKEN=... --transport stdio bright-data -- npx -y @brightdata/mcp

    Needs API_TOKEN

  • HyperbrowserOfficial · Hyperbrowser

    Scrape, crawl and extract with cloud browsers, plus browser-use and Claude computer-use agents.

    ★ 789pushed 2025-11-20MIT

    Local (stdio)
    claude mcp add hyperbrowser -- npx -y hyperbrowser-mcp
  • ScrapeGraphAIOfficial · ScrapeGraphAI

    Extract structured data from pages with a prompt, plus smart crawling.

    ★ 116pushed 2026-07-17MITon DevHunt

    Local (stdio)
    claude mcp add scrapegraphai -- uvx scrapegraph-mcp
  • OxylabsOfficial · Oxylabs

    Scrape any URL through Oxylabs proxies and return parsed content.

    ★ 104pushed 2026-10-15MIT

    Remote + local
    claude mcp add --transport http oxylabs https://mcp.oxylabs.io/mcp

    Headers: Authorization, X-Oxylabs-AI-Studio-Api-Key (add with --header where the client doesn't sign in with OAuth)

    Or run it locally: uvx oxylabs-mcp

  • SteelOfficial · Steel

    Browse the web with Steel's cloud browsers or a local Steel instance.

    ★ 58pushed 2026-10-20MIT

    See docs

    Setup is in the README (it needs a build step, an app plugin or your account details).

  • ScraplingCommunity

    Adaptive scraping with stealth HTTP requests and real browsers that get past common bot checks.

    ★ 89kpushed 2026-10-19BSD-3-Clause

    See docs

    Setup is in the README (it needs a build step, an app plugin or your account details).

  • SeleniumBaseCommunity

    Browser automation, testing and scraping on Selenium with a stealthy CDP mode.

    ★ 13kpushed 2026-10-19MIT

    See docs

    Setup is in the README (it needs a build step, an app plugin or your account details).

  • Playwright browser and API testing with screenshots and code generation.

    ★ 5.7kpushed 2025-12-13MIT

    Local (stdio)
    claude mcp add playwright -- npx -y @executeautomation/playwright-mcp-server

Using Cursor, VS Code or Claude Desktop? How to add an MCP server to each client. Read a server's permissions before you connect it to real accounts, and start with read-only or test keys.

FAQ

What is the best browser & web scraping MCP server?+

Start with the official server for the product you use: Browser Use, Chrome DevTools, Playwright, Skyvern, Apify. Popular community options: Scrapling, SeleniumBase, Playwright (ExecuteAutomation).

Which browser & web scraping MCP servers are official?+

11 of the 14 here are built by the vendor or the MCP project: Browser Use, Chrome DevTools, Playwright, Skyvern, Apify, Firecrawl, Bright Data, Hyperbrowser, ScrapeGraphAI, Oxylabs, Steel.

Which browser & web scraping MCP servers need no install?+

Skyvern, Apify, Firecrawl, Oxylabs run as hosted remote servers: add the URL to your client and sign in. The rest run locally with npx, uvx or Docker.

How do I add a browser & web scraping MCP server to Claude Code or Cursor?+

In Claude Code, run the command shown next to the server (claude mcp add ...). In Cursor, VS Code or Claude Desktop, add the same URL or command to the client’s mcp.json; the MCP servers hub has a snippet for each client.

More MCP servers: Databases · Dev tools & Git · Docs & code context · Cloud & DevOps · Monitoring & analytics · Security & code quality · Search & web data · Files & documents · Productivity & project management · Messaging & email · Design & frontend · Payments & commerce · AI, memory & reasoning · MCP tools launched on DevHunt