Apify Model Context Protocol (MCP) Server

Name: Apify Model Context Protocol (MCP) Server
Author: apify

@apify

About Apify Model Context Protocol (MCP) Server

The Apify MCP server enables your AI agents to extract data from social media, search engines, maps, e-commerce sites, or any other website using thousands of ready-made scrapers, crawlers, and automation tools available on the Apify Store.

Basic information

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "actors-mcp-server": {
      "command": "npx",
      "args": [
        "@apify/actors-mcp-server",
        "--tools",
        "actors,docs,apify/rag-web-browser"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is Apify Model Context Protocol (MCP) Server?

The Apify Model Context Protocol (MCP) server enables AI agents to extract data from social media, search engines, maps, e-commerce sites, and any other website using thousands of ready-made scrapers, crawlers, and automation tools from Apify Store. It supports OAuth for connecting from clients like Claude.ai or Visual Studio Code, and also supports agentic payments via x402 and Skyfire.

How to use Apify Model Context Protocol (MCP) Server?

You can connect via the hosted HTTPS endpoint at https://mcp.apify.com using OAuth or a Bearer token (Authorization: Bearer <APIFY_TOKEN>), or run locally via stdio with the command npx @apify/actors-mcp-server and the APIFY_TOKEN environment variable. For agentic payments, use the payment=x402 or payment=skyfire query parameter on the URL and follow the respective setup guides.

Key features of Apify Model Context Protocol (MCP) Server

Dynamically discover and use any Apify Actor as an MCP tool.
Pre-configured with apify/rag-web-browser and helper tools.
Supports OAuth for easy client integration.
Agentic payments with x402 (USDC on Base) and Skyfire (PAY tokens).
Streamable HTTP transport; compatible with Claude Desktop, Claude.ai, VS Code, Cursor, and others.
Loads each Actor’s input schema to create corresponding MCP tools.

Use cases of Apify Model Context Protocol (MCP) Server

Extract Facebook posts and comments from multiple pages or profiles.
Scrape Google Search Engine Results Pages (SERPs) for SEO or research.
Collect contact details from Google Maps listings.
Scrape Instagram posts, profiles, places, photos, and comments.
Use RAG Web Browser to search the web, scrape top URLs, and return content for AI analysis.

FAQ from Apify Model Context Protocol (MCP) Server

How does the Apify MCP Server differ from using Apify directly?

The server lets AI agents call any Apify Actor as a tool through the Model Context Protocol, dynamically discovering Actors and their input schemas without manual configuration.

What are the runtime requirements?

For the hosted server, only an MCP client is needed. For local stdio use, you need Node.js and an Apify API token. Agentic payment options require a wallet with USDC on Base (x402) or a Skyfire account.

Where is data processed and stored?

Data is processed in the Apify cloud. When using the hosted server at mcp.apify.com, all Actor runs happen on Apify’s infrastructure.

What authentication methods are supported?

You can authenticate via OAuth, a Bearer token (APIFY_TOKEN), or agentic payments using x402 (USDC) or Skyfire (PAY tokens) without an API token.

Can I use this server with any MCP client?

Yes, it is compatible with any MCP client. Tested clients include Claude Desktop, Claude.ai, ChatGPT, VS Code (Genie), Cursor, OpenCode, Kiro, and the Apify Tester MCP Client.