anycrawl-mcp-server

Anycrawl MCP Server

AnyCrawl connects web search, page extraction, and website crawling to MCP clients such as Cursor and Claude. It serves users gathering web content through AI tools, with controls for extraction formats, crawl scope, and job monitoring.

Server facts

anycrawl-mcp-server

Project

Author
any4ai
License
Not specified

Tech

Runtime / language
node
Transports
stdio
Remote available
No
Authentication
Not specified

Activity

GitHub stars
6
Last commit
Data as of

Tags

  • ai scrape
  • anycrawl
  • web crawler

Overview

AnyCrawl brings distinct workflows for known pages, broader website collection, and web discovery into MCP clients. The scrape tool accepts a target URL and can return Markdown, HTML, plain text, screenshots, or structured JSON. Extraction settings include tag filters, page wait times, and JSON schemas or prompts. Playwright, Cheerio, and Puppeteer are available as scraping engines. For discovery without a known URL, the search tool supports Google queries, language and country settings, pagination, and optional extraction of content from the returned pages.

Website collection uses the crawl tool, with controls for depth, page limits, domain boundaries, and path inclusion or exclusion. Separate tools let users inspect job progress and statistics, retrieve completed crawl content with pagination, or cancel an ongoing crawl. This separates content collection from monitoring and result retrieval. The project documents integrations with Cursor, VS Code, Windsurf, Claude Desktop, and Claude Code. Users can connect through the AnyCrawl cloud service or configure a self-hosted API instance, with API-key configuration described for both approaches.

anycrawl-mcp-server setup notes

  • anycrawl-mcp-server records stdio only; confirm a network transport upstream before planning remote access.
  • anycrawl-mcp-server has no recorded license; confirm reuse terms in its repository.

Related MCP servers