Maxun

Maxun is an open-source, self-hosted scraping platform that converts websites into structured data. It functions as a no-code platform for web scraping, crawling, and search, allowing users to automate data extraction without writing code. The software is deployed as a server-based web application and can be installed locally via Docker or without Docker, providing a flexible infrastructure for data collection.

Users interact with the platform through a point-and-click interface to create robots that emulate human browsing behavior. These robots can be managed through the web UI, a developer SDK, or a command-line interface for terminal-based control. The system supports various extraction methods, including a recorder mode for capturing manual actions and an AI mode for natural language descriptions, making it accessible for both non-technical users and developers.

Key features

• LLM-powered extraction using natural language descriptions
• Recorder mode to turn browsing actions into reusable robots
• Conversion of webpages into clean Markdown or HTML for AI workflows
• Automated website crawling with control over scope and discovery
• Automated web search with time-based filters
• Scheduling for recurring robot runs
• RESTful API endpoints generated from extracted website data
• Direct data export to Google Sheets and Airtable
• Support for authentication to extract data behind logins

Maxun integrates with the Model Context Protocol (MCP) and provides a developer SDK for programmatic scheduling and robot management. It is designed for use cases such as lead generation, market research, and content aggregation. The architecture allows for self-hosting to maintain full control over the scraping infrastructure and data privacy. By providing clean Markdown output, it specifically caters to users building AI agents or document processing pipelines that require structured web content.

Maxun is a unified platform for turning the web into structured APIs for developers and non-technical users.