Apify Mcp Server
Extract data from any website with thousands of scrapers, crawlers, and automations on Apify Store ⚡
750 servers in this category
MCPpedia tracks 750 browser MCP servers. 274 of them score at least 50 out of 100, and 162 clear 70 - about 22% of what has been published in this space. The remaining 476 are thinner: registry entries with no description, no published tool schema, or no commit in the last year. They are all still listed here, because knowing a server exists and is unmaintained is worth as much as knowing a good one exists.
Apify MCP Server leads the list at 95/100. Firecrawl MCP Server (95) and io.github.vmoranv/jshookmcp (95) follow. Every score is the same five weighted inputs: security (CVE scanning, tool-poisoning detection, and whether the server authenticates at all), maintenance (commit recency, release cadence, and download trend), documentation, client compatibility, and token efficiency - how much of your context window the tool definitions consume before you have asked anything.
9 of these servers are published by the vendor behind the underlying API rather than by a third party, which is usually the difference between an integration that tracks upstream changes and one that quietly stops working. 1 currently carry an open CVE; those are flagged on the server's own page with the advisory and its severity.
Also see: Best Browser MCP servers, Productivity, Developer Tools, Data, Finance
Last updated
Extract data from any website with thousands of scrapers, crawlers, and automations on Apify Store ⚡
Firecrawl, the context API for AI agents: search, scrape, interact with the web, plus paper search.
Search-first MCP server for JavaScript reverse engineering: 735 tools across 36 domains.
Native Safari browser automation for AI agents — 98 tools, zero Chrome overhead.
Claude in Chrome, reverse-engineered and open-source. No domain blocklist. Any Chromium browser. Same 18 MCP tools, same performance.
AI browser agent: browses, clicks, types, and reads real web pages from plain-English instructions.
Scrape, crawl, and map websites to Markdown or JSON via local CLI.
AI browser tools with Chrome CDP. Navigate, screenshot, interact. Multi-provider failover.
Web scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.
MCP server for automated web accessibility scanning with Playwright and Axe-core.
Evidence-bearing public webpage monitoring with hashes, diffs, signed webhooks, REST, and MCP.
Bright Data's Web MCP server enabling AI agents to search, extract & navigate the web
AI browser agent: browses, clicks, types, and reads real web pages from plain-English instructions.
Open-source web scraper for AI agents with scrape, crawl, and map tools
Self-hosted URL- and file-to-Markdown service for humans and AI agents - web pages, documents, images, audio, YouTube. PWA + REST + MCP + Claude Code skill, Reddit-aware, refreshable share links.
Read any page past its anti-bot wall, crawl sites, search without an API key, solve captchas locally
Quickly reads webpages and converts to markdown for fast, token efficient web scraping
Code-aware browser testing agent — 14 Playwright tools for AI code editors via MCP
Paid fallback for public webpages: clean text, Markdown, links, metadata, and JavaScript rendering.
MCP server for Chrome DevTools