The Apify MCP Server acts as a bridge, connecting AI clients to the Apify platform. It exposes Apify Actors as callable tools for AI applications, operating as a Model Context Protocol server. Developers can run it locally using Node.js and npm, or access a hosted service. This server enables AI agents to interact with Apify Actors, documentation, and storage, expanding their operational capabilities. It supports various AI clients and offers dynamic tool discovery.

What does Apify MCP Server connect to?

Apify MCP Server links directly to the Apify platform, integrating with Apify Actors, documentation, and storage. It functions as an MCP server, making Apify Actors available as callable tools for AI clients. For local development, it relies on Node.js v22 or higher and TypeScript templates. The server can be run locally via the npm package @apify/actors-mcp-server using "npx", or accessed as a hosted service at https://mcp.apify.com. Docker Hub integration is also supported for deployment.

Core features and tools

The server provides default callable tools including search-actors, fetch-actor-details, search-apify-docs, fetch-apify-docs, apify/rag-web-browser, and apify/web-fetch. It supports dynamic tool discovery, allowing AI agents to find and use any Apify Actor from the Apify Store. Additional helper tools manage Actor runs, access Apify storage (datasets, key-value stores), and manage Actor tasks. A report-problem tool is also available for issue reporting.

Who is it for?

Apify MCP Server is for AI agents, AI applications, LLMs, IDEs, CLIs, and developers building their own AI agents or coding assistants. It serves clients such as Claude.ai, Cursor, VS Code, ChatGPT, and JetBrains, along with any client adhering to the Model Context Protocol. It helps these systems perform web scraping, data extraction, and access Apify documentation. Skip this if your AI client does not adhere to the Model Context Protocol specification.

How to get started

The server authenticates using an API key passed as an environment variable or through browser-based OAuth for the hosted service. Users obtain an Apify account and an API token from the Apify Console. For local runs, set the APIFY_TOKEN environment variable, while the hosted service recommends Streamable HTTP as the transport protocol.

Use cases

• AI agents can use Apify Actors for web scraping and data extraction from various online sources.
• AI clients can dynamically discover and execute any Apify Actor from the Apify Store for on-demand task performance.
• AI coding assistants can search Apify documentation and manage data in Apify datasets and key-value stores.

Apify MCP Server connects AI clients to the Apify platform, making Apify Actors and data accessible as callable tools. It supports dynamic tool discovery and various authentication methods, including API tokens and OAuth. The server is compatible with many AI clients and IDEs, enabling advanced automation workflows. It offers a structured way for AI agents to interact with web data and Apify services. Try using the search-actors tool to find available Apify Actors.