---
title: Indexing Large Documents
description: Efficiently access huge documents without loading everything at once.
---
import { Aside } from '@astrojs/starlight/components';
# Indexing Large Documents
> *"It's not that I'm lazy, it's that I just don't care about loading 500 pages at once."*
For documents over 25,000 tokens, the indexing system enables on-demand fetching through MCP resources.
## Quick Example
```python
# Index once
result = await index_document("huge-novel.docx")
# Returns: {"doc_id": "abc123", "resources": {...}}
# Fetch chapters on demand
# chapter://abc123/1 → Chapter 1
# chapter://abc123/1.md → Chapter 1 as Markdown
```
---