Transform the Web into Structured Intelligence
✨ Turn any website into clean, contextualized data pipelines for your AI applications ✨
Maxun is the easiest way to extract web data with no code. The modern open-source alternative to BrowseAI, Octoparse and similar tools.
Go To App •
Documentation •
Website •
Discord •
Watch Tutorials
## What is Maxun?
Maxun helps you transform websites into structured APIs, clean markdown for AI workflows, and production-ready data pipelines — all in minutes.
### Ecosystem
1. **[Extract](https://docs.maxun.dev/category/extract)** – Emulate real user behavior and collect structured data from any website. No code required.
* **[Recorder Mode](https://docs.maxun.dev/robot/extract/robot-actions)** - Record your actions as you browse; Maxun turns them into a reusable extraction robot.
* **[AI Mode](https://docs.maxun.dev/robot/extract/llm-extraction)** - Describe what you want in natural language and let LLM-powered extraction do the rest.
2. **[Scrape](https://docs.maxun.dev/robot/scrape/scrape-robots)** – Convert full webpages into clean Markdown or HTML and capture screenshots. Ideal for AI workflows, agents, and document processing. No code required.
3. **[SDK](https://docs.maxun.dev/sdk/sdk-overview)** – A complete developer toolkit for scraping, extraction, scheduling, and end-to-end data automation.
Whether you prefer browsing through a website or integrating automation into your codebase, Maxun adapts to your workflow.
## How Does It Work?
Maxun uses web robots to power everything you can do on the platform. There are two types of robots, each designed for a different job.
### 1. Extract Robots
**Extract robots emulate real user behavior and capture structured data.**
Choose how to build them
### a. Recorder Mode: Record your actions as you browse
- Build robots visually by browsing like a human.
- Perfect for structured, deterministic data extraction.
### Example: Extract 10 Property Listings from Airbnb
[https://github.com/user-attachments/assets/recorder-mode-demo-video](https://github.com/user-attachments/assets/c6baa75f-b950-482c-8d26-8a8b6c5382c3)
### b. LLM Extraction (Beta): Describe what you want in plain language
- Use natural language to define extraction patterns.
- Works with closed source & open source LLMs.
Get Started with LLM Extraction: https://docs.maxun.dev/robot/extract/llm-extraction
### Example: Extract Names, Rating & Duration of Top 50 Movies from IMDb
https://github.com/user-attachments/assets/f714e860-58d6-44ed-bbcd-c9374b629384
### Core capabilities
- Extract from any website, including behind logins
- Convert sites into APIs, spreadsheets, and workflows
- Scale extractions and run on schedules or via API
- Handle infinite scrolling and pagination
- Auto-adapt to website layout & structural changes
### 2. Scrape Robots
**Built for clean content and AI workflows.**
- Get clean HTML and LLM-ready Markdown from any website
- Remove scripts, styling, ads, and clutter automatically
- Perfect for RAG systems, AI summarization, embeddings, and content pipelines
- Ideal for feeding clean data to LLMs
### Example: Scrape GitHub Trending Repositories in clean Markdown format
https://github.com/user-attachments/assets/c774cbd4-5a85-45b7-b41f-128ee570eae6
## Quick Start
### Getting Started
The simplest & fastest way to get started is to use the hosted version: https://app.maxun.dev. You can self-host if you prefer!
### Installation
Maxun can run locally with or without Docker
1. [Setup with Docker Compose](https://docs.maxun.dev/installation/docker)
2. [Setup without Docker](https://docs.maxun.dev/installation/local)
3. [Environment Variables](https://docs.maxun.dev/installation/environment_variables)
4. [SDK](https://github.com/getmaxun/node-sdk)
### Upgrading & Self Hosting
1. [Self Host Maxun With Docker & Portainer](https://docs.maxun.dev/self-host)
2. [Upgrade Maxun With Docker Compose Setup](https://docs.maxun.dev/installation/upgrade#upgrading-with-docker-compose)
3. [Upgrade Maxun Without Docker Compose Setup](https://docs.maxun.dev/installation/upgrade#upgrading-with-local-setup)
## Sponsors

LambdaTest
GenAI-powered Quality Engineering Platform that empowers teams to test intelligently, smarter, and ship faster.
|
## Features
- ✨ **Extract Data With No-Code** – Point and click interface
- ✨ **LLM-Powered Extraction** – Describe what you want; use LLMs to scrape structured data
- ✨ **Developer SDK** – Programmatic extraction, scheduling, and robot management
- ✨ **Handle Pagination & Scrolling** – Automatic navigation
- ✨ **Run Robots On Schedules** – Set it and forget it
- ✨ **Turn Websites to APIs** – RESTful endpoints from any site
- ✨ **Turn Websites to Spreadsheets** – Direct data export to Google Sheets & Airtable
- ✨ **Adapt To Website Layout Changes** – Auto-recovery from site updates
- ✨ **Extract Behind Login** – Handle authentication seamlessly
- ✨ **Integrations** – Connect with your favorite tools
- ✨ **MCP Support** – Model Context Protocol integration
- ✨ **LLM-Ready Data** – Clean Markdown for AI applications
- ✨ **Self-Hostable** – Full control over your infrastructure
- ✨ **Open Source** – Transparent and community-driven
## Use Cases
Maxun can be used for various use-cases, including lead generation, market research, content aggregation and more.
View use-cases in detail here: https://www.maxun.dev/#usecases
## Note
This project is in early stages of development. Your feedback is very important for us - we're actively working on improvements.
## License
This project is licensed under AGPLv3.
## Support Us
Star the repository, contribute if you love what we’re building, or [sponsor us](https://github.com/sponsors/amhsirak).
## Contributors
Thank you to the combined efforts of everyone who contributes!