Overview: Apify Model Context Protocol (MCP) Server

Title

Apify Model Context Protocol (MCP) Server

How to Use

The Apify MCP Server enables interaction with Apify Actors via two methods:

1. MCP Server Actor – An HTTP server accessible via Server-Sent Events (SSE).
2. MCP Server Stdio – A local server available via standard input/output (stdio).

To interact with the server, use compatible MCP clients such as:

• Claude Desktop (Stdio only)
• Visual Studio Code (Stdio & SSE)
• Apify Tester MCP Client (SSE with auth)

Key Features

• Integration with Apify Actors – Use any Apify Actor as a tool for tasks like web scraping, data extraction, and more.
• Pre-configured & Customizable – Default Actors include Instagram Scraper, RAG Web Browser, and Google Maps Email Extractor. Users can also define their own set of Actors.
• Helper Tools – Discover Actors (discover-actors), retrieve details (get-actor-details), and dynamically manage tools (add-actor-as-tool, remove-actor-from-tool).
• AI Agent Compatibility – Works seamlessly with AI frameworks implementing the MCP protocol.

Use Cases

• Web Scraping – Extract data from Facebook, Instagram, Google Maps, or SERPs.
• AI-Powered Automation – Enable AI agents to perform tasks like:
  • "Search the web and summarize recent trends about AI Agents."
  • "Find the top 10 Italian restaurants in San Francisco."
  • "Analyze The Rock’s Instagram profile."
• Dynamic Tool Management – Load or remove Actors on demand (supported by Tester MCP Client).

Example Actor Integrations

• Facebook Posts Scraper
• Google Maps Email Extractor
• RAG Web Browser

For more details, visit the Model Context Protocol website or explore the roadmap.

---

Note: The server currently supports Apify Actors and plans to integrate datasets and key-value stores in the future.