
Apify Actors
OfficialOfficial Apify MCP server — search and scrape the web via thousands of ready-made Actors; local or remote.
Add to your client
Copy the config for your MCP client and paste it into its config file.
npx -y @apify/actors-mcp-serverPaste into ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"apify-actors": {
"command": "npx",
"args": [
"-y",
"@apify/actors-mcp-server"
],
"env": {
"APIFY_TOKEN": "<your-apify-token>"
}
}
}
}Before you start
- An Apify account and API token from the Apify Console (Settings -> Integrations / API tokens)
- For local use: Node.js v22+ to run `npx @apify/actors-mcp-server`
- For remote use: an MCP client that supports remote servers and OAuth (e.g. Claude.ai, VS Code), or a Bearer token
- Note: running Actors consumes Apify platform credits/usage on your account
About Apify Actors
Apify's official MCP server connects AI agents to the Apify platform, where thousands of ready-made Actors (scrapers, crawlers, and automation tools) live in the Apify Store. Rather than writing scraping code, an agent can search the store, inspect an Actor's input schema, run it, and pull back structured results.
It supports dynamic tool discovery: search-actors finds relevant Actors, fetch-actor-details returns their input schema, and add-actor exposes a chosen Actor as its own callable tool — so the agent's toolset grows to match the task. call-actor executes Actors and get-dataset-items pages through their output datasets. A pre-configured apify/rag-web-browser Actor handles general web search and page reading out of the box.
It runs both ways: a hosted HTTP server at https://mcp.apify.com with OAuth (recommended, and where new features land first), and a local stdio server via the @apify/actors-mcp-server npm package. Authentication uses an Apify API token, supplied as a Bearer header remotely or the APIFY_TOKEN env var locally.
Tools & capabilities (11)
search-actorsDiscover Actors in the Apify Store relevant to a task.
fetch-actor-detailsReturn an Actor's specs, README, and input schema.
call-actorRun an Actor with input and retrieve its results.
add-actorDynamically expose a chosen Actor as its own MCP tool.
get-actor-runFetch information about a specific Actor run.
get-dataset-itemsRetrieve paginated items from an Actor run's output dataset.
get-key-value-store-recordRead a record from an Actor's key-value store.
abort-actor-runCancel a running Actor.
search-apify-docsSearch the Apify documentation.
fetch-apify-docsFetch a specific Apify documentation page.
apify/rag-web-browserPre-configured Actor tool for general web search and reading pages as content.
When to use it
- Use it when an agent needs to scrape a website but you don't want to write or host a scraper.
- Use it when you want the agent to search a marketplace of ready-made scrapers and pick the right one at runtime.
- Use it when extracting data from social media, search engines, maps, or e-commerce sites via prebuilt Actors.
- Use it when you need grounded web search/reading through the bundled rag-web-browser Actor.
- Use it when you want a fully hosted, OAuth-based scraping toolset rather than running anything locally.
Quick setup
- 1Create an Apify account and copy your API token from the Apify Console.
- 2For remote: add `https://mcp.apify.com` to your MCP client and authenticate via OAuth or an `Authorization: Bearer <APIFY_TOKEN>` header.
- 3For local: run `npx @apify/actors-mcp-server` with `APIFY_TOKEN` set in the environment.
- 4Restart the client so it discovers the Apify tools.
- 5Verify with search-actors, then fetch-actor-details and call-actor on a chosen Actor.
Security notes
The APIFY_TOKEN gates billable Actor runs; keep it secret and consider scoping exposed tools to limit what an agent can trigger. Actors execute remote scraping you may pay for, so review which Actors are enabled before running untrusted prompts.
Apify Actors FAQ
How do I authenticate?
With an Apify API token. The hosted server at mcp.apify.com supports OAuth, or you can pass Authorization: Bearer <APIFY_TOKEN>; locally set the APIFY_TOKEN environment variable.
Is it free?
The server itself is open source (MIT), but running Actors consumes Apify platform usage. New accounts get free monthly platform credits to start.
Should I use the hosted server or run it locally?
Apify recommends the hosted server at https://mcp.apify.com — it supports OAuth and gets the newest features (like output schema inference) first. Local stdio via npm is available for offline/embedded use.
What Node version do I need for local use?
Node.js v22 or higher to run `npx @apify/actors-mcp-server`.
How does the agent know which scraper to use?
It calls search-actors to find candidates, fetch-actor-details to read the input schema, and add-actor or call-actor to use one — tools are discovered dynamically.
Alternatives to Apify Actors
Official MCP reference server that fetches a URL and returns its content as clean Markdown, with chunking.
Official Firecrawl MCP server — scrape, crawl, map, search, and structured extraction for any LLM client.
Exa's neural web search and crawling MCP server — runs locally via npx or as a hosted remote endpoint.
Compare Apify Actors with: