Maxun

Active
GitHub TypeScript AGPL-3.0

Description

Maxun is an open-source no-code web data platform with Recorder Mode, AI Mode and full SDK support, turning any website into structured APIs, scheduled jobs and clean Markdown feeds for AI agents.

Key Features

  • Recorder mode — Record browser actions to auto-generate reusable extraction robots
  • LLM-powered extraction — Describe targets in natural language and let LLMs drive structured extraction
  • Multi-mode crawlers — Extract / Scrape / Crawl / Search robots cover the common data needs
  • Scheduling and API output — Cron scheduling, RESTful endpoints, Sheets and Airtable export
  • MCP integration — Built-in Model Context Protocol server, callable directly by Claude Desktop and other clients
  • Layout-change resilience — Auto-recovery when site structure changes mid-run

Use Cases

💡 Lead-generation bots scrape company directories and contact details by region
💡 Market-research agents aggregate competitor pricing, inventory and reviews across sites
💡 Continuously feed structured Markdown documents into RAG pipelines
💡 Monitoring agents track regulator, news, and social-sentiment changes on a cron schedule
💡 Handle data export from authenticated SaaS dashboards (CRM, support consoles)

Tags

web-scraping no-code playwright mcp data-extraction rpa

Categories

🌐 Browser Agent ⚡ Agent Tools

Quick Start

# Start the docker compose stack
git clone https://github.com/getmaxun/maxun
cd maxun
docker compose up -d

# Or use the SDK to create a robot directly
npm install @maxun/sdk
const { Maxun } = require('@maxun/sdk');
const client = new Maxun({ apiKey: 'YOUR_KEY' });
const robot = await client.createExtractRobot({
  url: 'https://example.com',
  selectors: { title: 'h1', price: '.price' }
});
const run = await robot.run();
Visit GitHub Visit Website View Docs

Related Projects