Apify MCP server
A hosted server that lets AI agents run ready-made web scrapers and automation tools from Apify Store.
- Stars
- 14.3k
- Forks
- 417
- Licence
- MIT
- Language
- TypeScript
- Last update
- 4 days ago
- Type
- AI model / agent
What it is
The Apify MCP server connects an AI agent to Apify Store, a marketplace of ready-made scrapers, crawlers, and automation tools called Actors. The agent can search the store, check an Actor's input and pricing, run it, and read the structured results. It can be used with many AI clients, either through a hosted endpoint or run locally over stdio.
- Ask an agent to search Apify Store for a scraper and run it on a target site
- Have an agent gather search engine results or web page content for monitoring
- Pull e-commerce or social media data into structured form through an agent
- Run Google Maps or web search Actors from within a chat client
Great for
- Developers who want an AI agent to pull data from websites without writing scrapers themselves
- Marketers or researchers who need search, map, or e-commerce data gathered by an agent
- Teams already using MCP-compatible clients such as Claude, Cursor, or VS Code
- People who want one connection instead of installing a separate MCP server per site
Not ideal for
- People who need to run everything offline or without an Apify account, since the README describes Apify sign-in and tokens for use
- Users who want a fully self-contained tool with no third-party service involved, because Actors run on Apify's platform
- Anyone who cannot review third-party scraper behavior or pricing before running it
How to get it
- 1
Use it in the browser (or self-host)
You'll need: Apify account (free to create, per README) · Node.js 22 or later (only for local stdio setup)
Add the hosted server to Claude Code
claude mcp add apify "https://mcp.apify.com" -t httpInstall config for Cursor via Apify CLI
apify mcp install cursor - 2
Run it
- Open the configurator at https://mcp.apify.com and pick your client
- Copy the generated configuration or command into your client
- Sign in to Apify in the browser window that opens
- Ask your agent to search Apify Store and run an Actor
- 3
Get the most out of it
- Choose only the tools you need in the configurator, since the README says the server loads only what each task needs
- Use the Apify CLI install commands to have credentials written for you
- Use the official Apify plugin for coding agents to get agent skills alongside the server
- Use the local npx server with an APIFY_TOKEN if you need a stdio setup
Downloads
Strengths
- The README describes one connection giving access to thousands of ready-made Actors
- Supports both a hosted server with OAuth and a local npx-based option
- Ships in several forms, including an npm package, a Docker catalog listing, and an MCPB bundle
Trade-offs
- Actor runs may incur costs; the README points to a Pricing and limits section but the details were not in the provided text
- Quality and behavior depend on each individual Actor from the store, which vary
- The hosted option requires signing in to Apify, and local use requires an Apify token
How it compares
| Project | Stars | Type | Level | Updated | Best for |
|---|---|---|---|---|---|
| apify-mcp-server (this) | 14.3k | AI model / agent | beginner | 4 days ago | Developers who want an AI agent to pull data from websites without writing scrapers themselves |
| context7 | 64.1k | AI model / agent | beginner | 4 days ago | Developers who use AI coding agents such as Claude, Cursor, or VS Code for libraries that change often |
| n8n-mcp | 23.2k | AI model / agent | intermediate | 4 days ago | People already using n8n who want an AI assistant to draft or check workflows |
| ruflo | 80.6k | Dev tool | intermediate | 4 days ago | Developers already using Claude Code or Codex who want multiple agents working together |
| ai-engineering-from-scratch | 76.3k | Learning resource | intermediate | 4 days ago | Students who want to understand how LLMs, retrieval, and agents work internally |
| codebase-memory-mcp | 48k | Dev tool | intermediate | 4 days ago | Developers who use an MCP-capable AI coding agent on a large codebase |
Stats synced from GitHub. The assessment above was drafted with AI from the project's README on 2026-10-29 and may miss recent changes — check the README before relying on a command. MIT licence: free to use, modify and distribute, including commercially, as long as you keep the copyright notice.