LibreAI / KI-Agenten / apify-mcp-server
// agents · apify/apify-mcp-server

apify-mcp-server

apify-mcp-server ist ein Open-Source-Projekt aus der Kategorie KI-Agenten von apify (apify/apify-mcp-server auf GitHub). Es hat 10.149 Sterne auf GitHub und 302 Forks, wird unter der Lizenz MIT veröffentlicht und wurde zuletzt am 8. Oktober 2026 auf GitHub aktualisiert.

MCP server that lets AI agents run thousands of Apify Store scrapers and crawlers as tools.

★ 10,1 Tsd. Sterne⑂ 302 ForksMITaktualisiert Okt 2026
▦ Visual — das README dieses Projekts enthält keinen einbettbaren Screenshot,
darum hier ein Platzhalter. Die Anleitung unten ist trotzdem vollständig.
// projektdaten

Projektdaten

GitHub-Sterne
10.149
Forks
302
Lizenz
MIT
Zuletzt auf GitHub aktualisiert
8. Oktober 2026
Kategorie
KI-Agenten
Eintragstyp
Vollständige Installationsanleitung
Offizielle Website
Official site
// so funktioniert es

So funktioniert es

  1. Choose hosted or local

    Use the hosted server at https://mcp.apify.com for the latest features and OAuth sign-in, or run it locally over stdio with npx for desktop clients.

  2. Connect your MCP client

    Add the server URL to clients like Claude.ai or VS Code (signing in via OAuth), or configure the stdio command with your Apify API token for tools like Claude Desktop.

  3. Let the agent discover Actors

    Ask for data in plain language; the agent searches the Store, inspects an Actor's details and pricing, and calls it with the right inputs.

  4. Collect structured results

    Read the Actor's output dataset from the run, and use the helper tools to inspect runs, logs, storage and schedules.

// funktionen

Warum es so beliebt ist

Any Actor becomes a tool

The server loads an Actor's input schema and exposes it as an MCP tool, so thousands of Store scrapers and automations are callable by agents.

Dynamic tool discovery

Helper tools let the agent search Actors, fetch their details and pricing, and call them on demand instead of you pre-configuring every scraper.

Sensible defaults

Two Actors come pre-configured - apify/rag-web-browser for web search plus content extraction, and apify/web-fetch for fetching pages as Markdown, text, HTML or links with JavaScript rendering.

Full platform access

Beyond running Actors, tools cover your runs and logs, datasets and key-value stores, saved tasks, schedules, builds and the Apify API reference itself.

Hosted with OAuth, or local stdio

Connect to mcp.apify.com with just the URL and an OAuth sign-in, or run npx @apify/actors-mcp-server locally with an APIFY_TOKEN.

Agentic payment options

For agents without an Apify account, the hosted server supports prepaid payment flows (AGI, x402 or Skyfire) to pay for Actor runs per use.

Advertisement
// selbst ausführen

Führe es selbst aus

  1. Clone the repository

    Clone the server source if you want to run or develop it locally; most users connect to the hosted endpoint or use npx instead.

    git clone https://github.com/apify/apify-mcp-server.git
    cd apify-mcp-server
  2. Add it to your MCP client

    Point your client at the hosted server URL and sign in with OAuth, or use the stdio package with your token.

    {"mcpServers": {"apify": {"url": "https://mcp.apify.com"}}}
  3. Restart and ask for data

    Restart your MCP client, then ask the agent to extract something - it will discover and call a suitable Actor.

    # e.g. "Scrape the latest posts from this public page" - see the README usage examples
// anwendungsfälle

Was du damit bauen kannst

Advertisement
// häufige fragen

Häufige Fragen

Who maintains the Apify MCP Server?

Apify maintains it as the official MCP interface to its platform and Store.

Is it free to use?

The server code is MIT-licensed and connecting is free, but Actor runs execute on the Apify platform and consume platform usage; many Actors are priced per run or per result by their authors, alongside Apify's own plan tiers.

Does it need a paid API key?

For local stdio use, yes - you set APIFY_TOKEN from your Apify account. The hosted server instead uses OAuth sign-in to your Apify account, and it also supports prepaid agentic payment options for agents without an account.

How does it compare with writing my own scrapers or a generic fetch tool?

A fetch tool only retrieves pages you name; this server gives agents a searchable store of thousands of maintained, site-specific scrapers with schemas, retries and structured output handled by the platform.

How do I get started?

Add https://mcp.apify.com to your MCP client, sign in with OAuth, and ask your agent for some web data - or try the Apify Tester MCP Client in the browser first.

// weiter entdecken

Ähnliche Tools

KI-Agenten

MobileBuildMCP

An MCP server and CLI that lets AI agents build, test and run iOS and macOS projects.

★ 6,5 Tsd. SterneMIT
Anleitung ansehen →
KI-Agenten

hexstrike-ai

MCP server letting AI agents run 150+ cybersecurity tools for authorized pentesting and research.

★ 12,5 Tsd. SterneMIT
Anleitung ansehen →
KI-Agenten

chrome-devtools-mcp

Chrome DevTools for coding agents: an MCP server that lets agents debug and inspect real Chrome

★ 53,1 Tsd. SterneApache-2.0
Anleitung ansehen →
KI-Agenten

n8n-mcp

An MCP server that lets Claude, Cursor and Windsurf build n8n workflows for you

★ 23,0 Tsd. SterneMIT
Anleitung ansehen →
KI-Agenten

klavis

An MCP integration platform that lets AI agents use tools reliably at any scale.

★ 5,8 Tsd. SterneApache-2.0
Anleitung ansehen →
KI-Agenten

jadx-ai-mcp

JADX plugin + MCP server that lets LLMs analyze Android APKs live — decompile, review, find vulnerabilities

★ 2,9 Tsd. SterneApache-2.0
Anleitung ansehen →
// alternativen

Alternativen im Bereich KI-Agenten

Die anderen Projekte der Kategorie KI-Agenten mit den meisten Sternen im LibreAI-Katalog — die realistische Auswahl, falls apify-mcp-server nicht das Richtige ist.

// mehr in agents

Mehr in KI-Agenten

Verwandte Projekte der Kategorie KI-Agenten im Katalog, die apify-mcp-server bei den Sternen am nächsten stehen — zum Weiterstöbern auf beiden Seiten.

Hol dir apify-mcp-server

Kostenlos und offen — hol dir den Quellcode, gib dem Repository einen Stern und leg los.