Apify MCP Server

Overview

The Apify MCP Server is the official Model Context Protocol (MCP) server developed by Apify. It exposes thousands of pre-built Actors (scrapers, crawlers, and automation tools) from the Apify Store as standardized MCP tools that AI agents can discover and use directly.

Hosted at https://mcp.apify.com, it allows seamless integration with clients like Claude Desktop, Claude.ai, Cursor, VS Code, and any MCP-compatible tool. No custom coding required—the AI can trigger real web data extraction, processing, and automation on Apify's robust cloud infrastructure.

Key Features

• Access to 5000+ Actors: Use ready-made tools for scraping LinkedIn, Instagram, TikTok, Google Maps, Amazon, YouTube, news sites, and virtually any website.
• Hosted & Easy Setup: Connect via OAuth using just the URL https://mcp.apify.com in your MCP client—no local server needed.
• Self-Hosted Option: Run locally with npx @apify/actors-mcp-server or Docker for full control.
• Structured Outputs: Automatic output schema inference helps AI agents understand and parse results reliably.
• Additional Tools: Actor management, dataset/key-value store access, run history, and custom tool configuration.
• Pay-per-Use: Billed according to standard Apify usage (free tier available with monthly credits).
• OAuth & Security: Secure authentication; supports streamable HTTP (SSE deprecated as of April 2026).

Use Cases

• Real-Time Web Research: Ask Claude to "scrape the latest job postings for AI engineers in Tokyo" and get structured results instantly.
• Social Media Monitoring: Extract posts, comments, or profiles from platforms like Instagram, X, or TikTok.
• Competitive Intelligence: Gather pricing, reviews, or product data from e-commerce sites.
• Lead Generation & Enrichment: Combine with other MCP servers (e.g., LinkedIn, Pappers) for powerful multi-tool workflows.
• Data Pipeline Automation: Let agents run crawlers, process data, and store results in Apify datasets.
• AI Agent Tooling: Turn generic LLMs into powerful web agents capable of browsing and acting on live internet data.

How It Works

1. Visit https://mcp.apify.com/ and configure your server (select Actors and tools).
2. Copy your unique MCP server URL or use the default hosted endpoint.
3. Add the server to your MCP client configuration (Claude Desktop, Cursor, etc.).
   • Hosted example: https://mcp.apify.com
   • Local: npx @apify/actors-mcp-server --actors=actor1,actor2
4. The AI automatically discovers available tools and calls them with natural language instructions.
5. Results return structured into the conversation, with full traceability.

Repository: github.com/apify/apify-mcp-server

Getting Started

• Hosted (Recommended): Sign in at mcp.apify.com with your Apify account and configure.
• Local Run: Requires Node.js and an Apify API token.
• Documentation: Full guides available in the Apify docs and GitHub repository.

The Apify MCP Server is one of the most powerful and comprehensive web data tools in the MCP ecosystem, turning any compatible AI into a capable web automation agent.

Important: Always respect website terms of service and robots.txt when scraping. Apify provides proxy and anti-blocking infrastructure to help with reliable data collection.