What it does
The Apify MCP Server exposes thousands of Apify Actors as tools for AI agents to use. Actors are web scrapers, crawlers, and automation tools that extract data from social media platforms, search engines, maps, e-commerce sites, and general websites. The server supports OAuth for cloud-hosted access via https://mcp.apify.com and token-based authentication for local stdio deployments.
Who it's for
Backend engineers and AI agent developers building systems that need web data extraction. Ideal for teams integrating Claude or other AI assistants into data pipelines, research workflows, or automation systems requiring real-time web scraping and information retrieval.
Common use cases
- Extract posts and engagement data from Facebook, Instagram, or other social media
- Scrape Google Search results (SERPs) for SEO research or data collection
- Extract contact details and business information from Google Maps
- Perform web searches and scrape top results for retrieval-augmented generation (RAG)
- Automate data collection from e-commerce sites or general websites
Setup pitfalls
- Requires
APIFY_TOKENenvironment variable for local deployments; without it, authentication will fail - Package contains 4 detected secrets — do not commit tokens to version control
- Network access required for all scraping operations; ensure firewall rules permit outbound HTTPS to target websites
- Filesystem write access needed for data storage and results — run in isolated sandbox; legacy SSE transport removed — use either the hosted endpoint or
stdio, not the old/ssesuffix