Apify MCP Server is a Model Context Protocol server that enables AI agents to extract data from websites using ready-made scrapers, crawlers, and automation tools.
The tool solves the problem of integrating web data extraction into AI agent workflows. It provides access to thousands of pre-built actors available on the Apify Store, which can scrape social media, search engines, maps, e-commerce sites, and other websites. Rather than building custom scrapers, agents can invoke these existing tools through the MCP interface. The server supports OAuth authentication, allowing connection from AI clients like Claude.ai or Visual Studio Code using just a URL. A hosted version at mcp.apify.com is recommended for the best experience, as it includes features like output schema inference for structured results that are not available in local stdio deployments.
Adoption is most suitable for developers building AI agents that need reliable web data extraction capabilities without implementing scrapers from scratch. The tool works with any MCP-compatible client, including Claude Code, Claude.ai, Cursor, and VS Code. Teams should choose this if they want to leverage Apify's existing actor ecosystem rather than maintain custom scraping infrastructure. The project supports multiple payment models for running actors, including agentic payments through AGI tokens, direct per-request payments for specific actors, and Skyfire integration.
The project maintains active development with a build status pipeline and regular updates to the codebase. The tool has undergone significant evolution, including removal of legacy SSE transport in favor of Streamable HTTP, indicating ongoing refinement of its core architecture. The project is distributed as an npm package and actively maintained with compatibility across multiple AI development environments.
