MCP servers for browser automation and web scraping
Browser servers are some of the most used MCP servers, because so much work happens on web pages. They let an assistant open a page, click, type, read the result and take screenshots, or fetch and clean a page's content for the model to read. Developers use them to test and debug; everyone else uses them to research and collect. This list ranks them by GitHub stars.
Automation, debugging and scraping
Automation servers drive a real browser through Playwright or Chrome's DevTools protocol. They suit testing a web app, reproducing a bug or doing a chore on a site with no API.
Scraping and search servers skip the clicking and return a page's text, or a set of search results, ready for the model. They are faster and cheaper per page, and they fail on sites that need a login or block bots.
For testing, prefer servers that read the structure of a page over ones that work from screenshots. They are faster and break less when the layout shifts.
Use them responsibly
Respect each site's terms and robots rules, and keep request rates low. Never point an automated browser at accounts you are not allowed to automate.
A logged-in browser is powerful. When an assistant drives one, it can do anything you can on those sites, so use a separate profile and watch the first runs.
SkillScout's copy of the official MCP Registry had 1,021 servers that mention browsing, scraping or crawling on 27 October 2026.
The list
From SkillScout's catalogue, counted on 27 October 2026: 81 entries matched this topic, and these are the 12 with the most GitHub stars, one per project, after a review for false matches. Descriptions are each project's own. Stars show popularity, not fit.
-
Scrapling
An adaptive Web Scraping framework that handles everything from a single request to a full-scale crawl!
Get the code, then follow its README:
git clone https://github.com/D4Vinci/Scrapling -
chrome-devtools-mcp
Chrome DevTools for coding agents
Get the code, then follow its README:
git clone https://github.com/ChromeDevTools/chrome-devtools-mcp -
playwright-mcp
Playwright MCP server
Get the code, then follow its README:
git clone https://github.com/microsoft/playwright-mcp -
browser-tools-mcp
Monitor browser logs directly from Cursor and other MCP compatible IDEs.
Get the code, then follow its README:
git clone https://github.com/AgentDeskAI/browser-tools-mcp -
firecrawl-mcp-server
Official Firecrawl MCP Server - Adds powerful web scraping and search to Cursor, Claude and any other LLM clients.
Add it to Claude Code (from the official MCP Registry):
claude mcp add firecrawl-mcp-server -- npx -y firecrawl-mcp -
exa-mcp-server
Exa MCP for web search and web crawling!
Get the code, then follow its README:
git clone https://github.com/exa-labs/exa-mcp-server -
mcp-server-browserbase
Allow LLMs to control a browser with Browserbase and Stagehand
Add it to Claude Code (from the official MCP Registry):
claude mcp add mcp-server-browserbase -- npx -y @browserbasehq/mcp-server-browserbase -
ddgs
A metasearch library that aggregates results from diverse web search services
Get the code, then follow its README:
git clone https://github.com/deedy5/ddgs -
tavily-mcp
Production ready MCP server with real-time search, extract, map & crawl.
Get the code, then follow its README:
git clone https://github.com/tavily-ai/tavily-mcp -
anysearch-mcp-server
Unified real-time search MCP server supporting general web search, vertical domain search, parallel batch search, and full-page URL content extraction.
Get the code, then follow its README:
git clone https://github.com/anysearch-ai/anysearch-mcp-server -
open-webSearch
Multi-engine MCP server, CLI, and local daemon for agent web search and content retrieval, skill-guided workflows, no API keys.
Get the code, then follow its README:
git clone https://github.com/Aas-ee/open-webSearch -
stealth-browser-mcp
The only browser automation that bypasses anti-bot systems. AI writes network hooks, clones UIs pixel-perfect via simple chat.
Get the code, then follow its README:
git clone https://github.com/vibheksoni/stealth-browser-mcp
Questions
What is the Playwright MCP server?
Microsoft's MCP server that lets an assistant control a browser through Playwright, using the page's accessibility tree rather than screenshots, which makes it fast and reliable.
Can an MCP server scrape websites?
Yes. Scraping servers fetch pages and return clean text or structured data. Check the site's terms first, and expect sites with strong bot protection to block some requests.
Related
- Best AI tools, sorted by role and task
- Best MCP serversThe most-starred MCP servers, with the install line where the registry has it.
- Best AI tools for developersMCP servers and skills that give a coding agent your repo, docs and tools.
- Best AI browser extensionsOpen-source Chrome extensions that put a model next to the page.
Your own list
This page ranks by popularity. SkillScout's scan ranks by fit: the AI you already use writes a short profile of how you work, and SkillScout scores about 22,000 AI tools against your gaps and hands back a short list with install steps. It is free, and the server never saves your profile.