> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://cla-demo.postmansolutions.com/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://cla-demo.postmansolutions.com/_mcp/server.

# Overview

In Fluxx, users upload documents that may contain **multiple forms bundled into a single PDF**. Before any data extraction can happen, the platform needs to know:

1. What forms are inside the PDF?
2. Which pages belong to which form?

A separate **classifier** service answers those questions. Once it does, the **document splitter** takes over, it physically separates each form into its own PDF file so that downstream extractors can process each form independently.

## Summary

> The document splitter listens for classified-document events, downloads the PDF and its classification metadata, splits the PDF into per-form chunks, uploads each chunk, and publishes events that tell the rest of the platform where those chunks live.

## What It Is NOT

* It does **not** classify documents, that is done by a separate classifier service upstream.
* It does **not** extract data from forms, that is done by extractor services downstream.
* It does **not** expose a public REST API for triggering splits, it is purely event-driven.