Model Context Protocol Server that exposes Apify Actors as MCP tools (HTTP-SSE or stdio).
https://github.com/apify/actors-mcp-serverStop building custom scrapers from scratch. The Apify Actors MCP Server instantly connects your AI assistant to the entire Apify ecosystem - over 5,000 pre-built web scraping and automation tools that handle everything from social media extraction to contact discovery.
You need data from Instagram, Google Maps, LinkedIn, or any other platform. Your options are usually:
Meanwhile, there's an entire marketplace of battle-tested, constantly maintained scrapers just sitting there.
The Apify Actors MCP Server bridges your AI assistant (Claude Desktop, VS Code, etc.) with Apify's massive actor library. Instead of asking "Can you scrape Instagram?" and getting a generic response, you can ask:
"Find and analyze The Rock's Instagram engagement rates" - and your AI will automatically use the Instagram Scraper actor, extract the data, and perform the analysis.
"Get contact details for Italian restaurants in San Francisco" - it'll use the Google Maps scraper with contact extraction, no API keys or rate limits to worry about.
"Research recent AI agent trends and summarize findings" - the RAG Web Browser actor will search, scrape multiple sources, and return structured content ready for analysis.
Here's where it gets interesting - your AI can discover and use new tools on the fly. Need to scrape a platform you've never worked with before? Your assistant can:
search-actors)get-actor-details)add-actor)No more "I can't help with that" responses. Your AI becomes genuinely useful for data extraction tasks.
Social Media Intelligence
"Analyze competitor engagement on Instagram and Facebook"
Your AI automatically uses multiple actors (Instagram Scraper, Facebook Posts Scraper) to gather data, then provides competitive analysis.
Lead Generation
"Find email addresses for SaaS companies in Austin with 10-50 employees"
Combines Google Maps extraction with LinkedIn data to build targeted prospect lists.
Market Research
"What are people saying about ChatGPT on Reddit this week?"
Uses Reddit scrapers to collect recent discussions, sentiment analysis included.
Local Development (Stdio) Perfect for Claude Desktop integration. One npm command and you're running:
{
"mcpServers": {
"actors-mcp-server": {
"command": "npx",
"args": ["-y", "@apify/actors-mcp-server"],
"env": {"APIFY_TOKEN": "your-token"}
}
}
}
Production Deployment (HTTP/SSE)
Run as a standby actor on Apify's platform. Your AI connects via HTTP, and the server scales automatically. No infrastructure headaches.
Beyond just running scrapers, you get full actor lifecycle management:
get-actor-run, get-actor-log)get-dataset-items)get-key-value-store-record)abort-actor-run)Your AI becomes a complete data extraction orchestrator.
VS Code: One-click install buttons for immediate setup
Claude Desktop: Add 5 lines to your config file
Custom Clients: Standard MCP protocol, works with any compliant client
The server handles all the complexity - actor schema parsing, input validation, result formatting. You just ask questions and get structured data back.
Instead of spending days building scrapers that break, you get immediate access to tools that:
Your AI goes from "I can't access external websites" to "Let me extract exactly what you need from any platform."
The actor marketplace has tools for virtually every major platform - Google, Facebook, Instagram, LinkedIn, Amazon, Airbnb, TikTok, Reddit, Twitter, and hundreds more. Plus specialized tools for contact extraction, SEO analysis, price monitoring, and data enrichment.
npx -y @apify/actors-mcp-serverNo complex setup, no infrastructure management. Just immediate access to industrial-grade web scraping capabilities through your existing AI workflow.
The server is open source (Apache 2.0), actively maintained by Apify, and already trusted by hundreds of developers who've contributed 236 stars and 27 forks.
Stop building scrapers. Start using the ones that already work.