Back

Apify MCP Server

Connect to the Apify MCP server to search the Apify Store, run Actors, track runs, and pull scraped datasets using AI agents on Gumloop, Claude, or Cursor.

MCP server URL

https://mcp.gumloop.com/apify/mcp

Works with

Tools 11

  • Search Actors

    Search runnable Actors in the Apify Store

    3 credits

  • Get Actor

    Get Actor metadata and optionally its input schema. Full-permission Actors cannot run with the Gumloop-managed Apify key; connect your own Apify API key for those Actors.

    3 credits

  • Validate Actor Input

    Validate input for an Actor build

    3 credits

  • Run Actor

    Run an Actor asynchronously or synchronously. The Gumloop-managed Apify key only supports execution_mode="sync" for public limited-permission Actors. Async runs require the user's own Apify API key.

    3 credits +

  • List Tasks

    List saved Apify Actor tasks. Requires the user's own Apify API key.

    3 credits

  • Get Task

    Get a saved Apify Actor task. Requires the user's own Apify API key.

    3 credits

  • Run Task

    Run a saved Apify Actor task asynchronously or synchronously. Requires the user's own Apify API key.

    3 credits +

  • Get Run

    Get Actor run status and metadata. Requires the user's own Apify API key.

    3 credits

  • Abort Run

    Abort an Actor run. Requires the user's own Apify API key.

    3 credits

  • Get Run Log

    Get an Actor run log. Requires the user's own Apify API key.

    3 credits

  • Get Dataset Items

    Get items from an Apify dataset. Requires the user's own Apify API key.

    3 credits +

Frequently asked questions

Related MCP servers

Firecrawl

Search, scrape, and map websites using Firecrawl's web data API

  • Developer Tools
  • OAuth
Connect

Parallel

Extract content from web URLs using Parallel AI

  • Developer Tools
  • API key
Connect

Exa

Interact with Exa.ai API for web search - intelligently search the web, retrieve content, find similar pages, get LLM-powered answers, and conduct deep research

  • Developer Tools
  • API key
Connect

Google Maps

Server for scraping Google Maps data using Apify

  • Google Workspace
  • OAuth or API key
Connect

Google Sheets

Read, write and update your Google Sheets

  • Google Workspace
  • OAuth
Connect

Slack

Interact with Slack channels and messages

  • Productivity & Collaboration
  • OAuth
Connect

Further reading

What is Apify MCP?

The Apify MCP server gives AI agents access to your Apify web scraping and automation stack. That means agents can search the Apify Store for runnable Actors, read an Actor’s metadata and input schema, validate inputs before a run, start Actors synchronously or asynchronously, run your saved Actor tasks, track run status and logs, abort runs in progress, and pull the items an Actor scraped into its dataset. It works across the core Apify objects an agent needs: Actors, tasks, runs, and datasets.

If your team writes one-off scripts to kick off scrapers, babysits the Apify console waiting for runs to finish, then copies results out of datasets by hand, an AI agent can take over that loop. Describe what you need, and your AI agent will handle the scraping and data retrieval for you.

MCP stands for Model Context Protocol. It’s an open standard that gives AI agents a way to connect to external tools and services. Instead of writing code against the Apify API, polling run status, handling pagination, and parsing dataset JSON yourself, you connect Apify to Gumloop once. After that, you can run scrapers and collect their results just by chatting with your AI agent.

What you can do with Apify MCP on Gumloop

  • Search the Apify Store for the right Actor

    Use a query to find runnable Actors in the public Apify Store, so an agent can pick the scraper or automation that fits the job instead of you browsing the catalog.

  • Inspect an Actor before you run it

    Pull an Actor’s metadata and its input schema, then validate a proposed input against that schema. Your agent confirms the run is set up correctly before spending a single credit.

  • Run Actors to scrape and automate

    Start any Actor synchronously or asynchronously with the inputs you specify. An agent can launch a web scraper, a crawler, or an automation Actor and wait for results or kick it off in the background.

  • Run your saved Actor tasks

    List your saved tasks, read a task’s configuration, and run a task on demand. Reuse the scraper setups your team already saved in Apify without rebuilding the inputs each time.

  • Track run status and read logs

    Get the status and metadata for any run, pull its log to see what happened, and abort a run that is going wrong. Your agent keeps tabs on long jobs so nobody has to refresh the console.

  • Pull scraped data from datasets

    Fetch the items an Actor wrote to its dataset and hand them straight to an AI agent to clean, summarize, or route. The scraped results land where the rest of your automation can use them.

How to connect the Gumloop Apify MCP Server

  1. Create a Gumloop account

    Sign up at gumloop.com. Every new account starts with a 14-day free trial.

  2. Add the Apify MCP server

    Copy your MCP server URL from Gumloop and add it to your preferred client (Claude, Cursor, or Gumloop workflows). You'll authorize on first use.

  3. Start using Apify in your AI workflows

    That's it. Your AI agent can now search the Store, run Actors and tasks, track runs, and pull scraped datasets. Use it inside a Gumloop automation, in Claude Desktop, or in Cursor.

Apify MCP use cases

  • On-demand web scraping for growth teams

    A growth team wants fresh leads from a directory site. A Gumloop agent searches the Apify Store for a matching scraper, validates the input, runs the Actor, then pulls the dataset items and writes them to Google Sheets for the team to work. No script to maintain, no console to watch.

  • Competitive and price monitoring for operations teams

    An ops team tracks competitor pages and listings. An agent runs a saved Apify task on a schedule, fetches the dataset of scraped prices, compares it against last week’s pull, and posts the deltas to a Slack channel. The team sees what changed without exporting a single CSV.

  • Pipeline-ready datasets for data teams

    A data team needs scraped results in a usable shape. An agent runs the right Actor, reads the dataset items, has an LLM normalize and dedupe the records, then loads the clean rows into Google Sheets or another destination. The handoff from raw scrape to analysis-ready data happens in one agent run.

  • Market and lead research for research teams

    A research team gathers signals across many sites. An agent chains an Apify Actor with web search and enrichment tools, runs the scrape, pulls the dataset, and compiles a sourced brief. Researchers get a finished summary instead of a stack of raw pages.

  • Catalog and review scraping for ecommerce teams

    An ecommerce team wants competitor catalogs and reviews. An agent runs a product or review Actor, tracks the run to completion, fetches the dataset items, and drops a structured rollup into a shared sheet. The team watches the merchandising picture without manual collection.

Why use Gumloop for Apify MCP

  • Start with no token, add yours when you need full access

    Most Apify MCP servers you’ll find on GitHub make you store an API token in environment variables and write code to handle auth, run polling, and pagination. With Gumloop you can search the Store and run public limited-permission Actors synchronously on the built-in connection right away, then add your own Apify API token once, stored securely, to unlock your saved tasks, run management, datasets, async runs, and private Actors. No config files, token juggling, or coding necessary.

  • Works with multiple MCP clients

    Use the Apify MCP server endpoint in Claude Desktop, Cursor, or directly inside Gumloop agents. Same server URL, works with any MCP client.

  • Chain Apify with 100+ other integrations

    Combine Apify with Google Sheets, Slack, Gmail, web search, and other MCP tools in a single AI agent. An agent can run a scraper, pull the dataset, process it with an LLM, and write the result to a sheet or a Slack message.

  • Enterprise-grade and scalable

    Built for teams, with role-based permissions and dedicated support for Pro users. Run a single scraper or coordinate many across a large org. For details on Gumloop’s security practices, see trust.gumloop.com.

  • Start with a free trial

    You can test the Apify MCP integration during Gumloop’s 14-day free trial before committing. Paid plans start at $37/month.