Use Apify MCP for agent web extraction
Use the Apify MCP server to let MCP-compatible agents run Apify Actors for structured extraction from search, maps, ecommerce, social platforms, and arbitrary websites.
npx skills add agentskillexchange/skills --skill use-apify-mcp-for-agent-web-extraction
Use Apify MCP when an operator wants an agent to call production scrapers and crawlers as MCP tools instead of writing scraping code during the task. The agent/operator connects an MCP client to the hosted https://mcp.apify.com endpoint with OAuth or an Apify token, or runs the local stdio server with npx @apify/actors-mcp-server, then lets the agent select and run Actors for SERPs, maps, ecommerce, social data, RAG browsing, and custom site extraction. Invoke it instead of using Apify manually when an agent needs to choose, parameterize, execute, and consume Actor outputs inside a larger research or data pipeline. The scope boundary is MCP-mediated Actor execution from agent clients; it is not a generic Apify platform, Crawlee SDK, or web-scraping-library listing.
What this skill actually does
Inputs and prerequisites: MCP-compatible client such as Claude Desktop, Claude.ai, VS Code, Cursor, or Claude Code; hosted Apify MCP endpoint or @apify/actors-mcp-server; Apify account or API token where required; selected Apify Actors.
Setup notes: For hosted use, configure the MCP client with https://mcp.apify.com and authenticate with OAuth or an Authorization bearer token. For local stdio use, set APIFY_TOKEN and configure the MCP client server command as npx @apify/actors-mcp-server.
Source and verification boundary: use https://docs.apify.com/platform/integrations/mcp as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a MCP workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.