Apify MCP server
Official Apify server that exposes 6,000+ Actors plus run, dataset, and store tools so agents can scrape and automate the web.
Apify MCP is the official server from Apify, a cloud platform of pre-built scrapers and automations called Actors. It lets an agent discover Actors in the Apify Store, inspect their input schemas, and call them to scrape websites or run automation jobs, then read back the results from the run's dataset or key-value store. Rather than hard-coding a fixed list of tools, the server can dynamically expose any Actor as a callable tool, so a single connection reaches the whole catalog of more than six thousand ready-made scrapers.
It installs two ways. Locally it runs over stdio with npx using the @apify/actors-mcp-server package and an APIFY_TOKEN env var. Apify also hosts a remote Streamable HTTP endpoint at https://mcp.apify.com that supports OAuth for clients like Claude and VS Code, as well as bearer-token auth with the same APIFY_TOKEN. Beyond the Actor-as-tool surface, it ships helper tools for searching Actors, fetching their details, managing runs, reading datasets and key-value stores, and querying Apify's own documentation.
Quick install
Copy-paste configs are provided for all 8 supported clients. Pick your client below.
Add to ~/.claude.json
{
"mcpServers": {
"apify": {
"type": "http",
"url": "https://mcp.apify.com"
}
}
}claude mcp add --transport http apify https://mcp.apify.comHeads up
- First tool call opens a browser to authorize.
Available tools
| Tool | Description |
|---|---|
| search-actors | Search the Apify Store to find Actors that match a task. |
| fetch-actor-details | Fetch an Actor's metadata, including its input schema and documentation. |
| call-actor | Run an Actor with the given input and return its results. |
| add-actor | Dynamically register an Actor as a dedicated callable tool for the session. |
| get-actor-run | Retrieve the details and status of a specific Actor run. |
| get-actor-run-list | List Actor runs, optionally filtered by status. |
| get-actor-log | Read the execution log of an Actor run. |
| get-actor-output | Retrieve the complete output produced by an Actor run. |
| abort-actor-run | Stop a running Actor execution. |
| get-dataset | Get metadata about a dataset. |
| get-dataset-items | Retrieve items from a dataset with optional filtering and pagination. |
| get-dataset-schema | Generate a JSON schema inferred from a dataset's items. |
| get-dataset-list | List the datasets in your Apify account. |
| get-key-value-store | Get metadata about a key-value store. |
| get-key-value-store-keys | List the keys held in a key-value store. |
| get-key-value-store-record | Fetch a single record from a key-value store by key. |
| get-key-value-store-list | List the key-value stores in your Apify account. |
| search-apify-docs | Search Apify's documentation for relevant pages. |
| fetch-apify-docs | Retrieve the full content of an Apify documentation page. |
Required configuration
- APIFY_TOKENRequired
Apify API token used to authenticate and run Actors. Required for the local stdio install and for bearer auth against the remote endpoint.
What you can do with it
Scrape any site with a ready-made Actor
Search the Apify Store for a scraper, add it as a tool, and call it with a URL to get structured results without writing scraping code.
Drive long-running jobs from an agent
Kick off an Actor run, poll its status and logs, then read items from the resulting dataset or key-value store once it completes.
FAQ
- Is it free?
- The server is open source and free to run; calling Actors consumes Apify platform usage, which requires an APIFY_TOKEN. Apify provides a free tier with monthly platform credits to get started.
- Does it support remote/OAuth?
- Yes. Apify hosts a remote Streamable HTTP endpoint at https://mcp.apify.com that supports OAuth for clients like Claude and VS Code, plus bearer-token auth using your APIFY_TOKEN. The local stdio package uses the token via an env var.