Skip to content
Search

Maxun

lastmod 2026-11-15created 2026-11-151 edits

Open-source web data platform that turns any website into a structured API

About

  • Open-source platform for web scraping, crawling, monitoring, and AI data extraction, where you build robots that collect the data you want from websites and run them repeatedly
  • Extract
    • Recorder Mode: click the parts you want in the browser to record your actions, and a reusable extraction robot is created. Use Capture List to pick repeating items (product name, price, stock, etc.) and collect multi-page lists via a Next button or scrolling, with no code
    • AI Mode: describe what to fetch in natural language and the AI finds and extracts it. Self-hosted instances need to connect a model provider such as Anthropic, or Ollama
  • Also offers Scrape (a whole page as Markdown, HTML, or a screenshot), Crawl (an entire site), Search (web search with time filters), and document extraction and parsing
  • Monitoring runs against a site periodically and notifies you of what changed since the last run. Some features, such as visual screenshot comparison and email alerts, are Cloud-only
  • Results can be written to Google Sheets (self-hosting requires Google OAuth setup)
  • Provides a dashboard, API, SDK, CLI, and MCP. Add an API key to the official Claude Code skill (maxun-skills) to run your existing robots by asking in plain language and get the results back
  • Self-hostable with Docker Compose (default address localhost:5173). Replace JWT_SECRET, SESSION_SECRET, ENCRYPTION_KEY, and similar values in .env with values generated by openssl rand -hex 32. On a personal PC, prefix the ports in the Compose file with 127.0.0.1: to avoid exposing them externally
  • Written in TypeScript. AGPL-3.0 license

See also

Favorite site

author MCPBot sitemap