Your entire DatoCMS docs in one file: meet llms-full.txt

TL;DR

• We've shipped llms-full.txt — our complete documentation, all 500+ pages, in one clean Markdown file optimized for AI tools.

• Drop it into Claude Projects, custom GPTs, NotebookLM, Cursor, or any AI assistant and get accurate, context-aware answers about DatoCMS without hunting through docs.

• It's the natural next step after our LLM-ready documentation export — now you get the entire knowledge base in one shot!

Why this matters

A few weeks ago, we added the ability to export any docs page as Markdown with a single click. It works beautifully for single pages. But when you're building something real with DatoCMS, you don't just need one page. You need context. You need to understand how the pieces fit together.

Here's the thing about working with AI tools: they're only as good as the context you give them. Ask ChatGPT to help you build a content migration script without any docs? You'll get generic code that might corrupt your data. Paste in the full migration documentation? You get production-ready scripts with proper error handling and validation that actually work with your content model.

Before llms-full.txt, feeding complete context meant opening 10+ tabs, copy-pasting page after page, and hoping the AI could piece it together.

Now? One file. Full context. Every time.

That's llms-full.txt — our entire documentation compiled into a single, perfectly formatted Markdown file. API references, guides, migrations, plugins, CMA, CDA, all of it. One URL. One paste. Complete context.

Here's how to actually use it

Claude Projects: Your DatoCMS expert on demand

Claude Projects let you attach custom knowledge to Claude. This is where llms-full.txt absolutely shines.

Setup (takes 30 seconds):

1. Go to claude.ai and create a new Project

2. Drop the llms-full.txt file and use these instructions as your starting point

3. Give it a name like "DatoCMS Docs"

Now you can:

• Ask "How do I migrate from WordPress to DatoCMS?" and get step-by-step instructions

• Say "Write a script to bulk-update all my blog posts" and get working TypeScript

• Plan complex content model migrations with full awareness of DatoCMS capabilities

Claude remembers everything from llms-full.txt across every conversation in that Project. It's like having a DatoCMS architect in your back pocket.

Custom GPTs: Build your own DatoCMS assistant

Want a ChatGPT that only talks DatoCMS? You can build one in minutes.

How to build it:

1. Go to ChatGPT GPT Builder

2. Click "Create a GPT"

3. In the Knowledge section, upload llms-full.txt (download it first from here)

4. Give it instructions like: "You are a DatoCMS expert. Answer questions using only the provided documentation. Include code examples when relevant." You can use these instructions as your starting point.

Pro tip: Check out our official DatoCMS Expert GPT to see what's possible!

NotebookLM: Research and learn DatoCMS

Google's NotebookLM is brilliant for deep research and learning.

Setup:

1. Create a new notebook in NotebookLM

2. Add a source → paste https://www.datocms.com/docs/llms-full.txt

3. Let it process (~30 seconds)

Now you can:

• Ask "Compare the CMA and CDA APIs — when should I use each?"

• Say "Create a study guide for learning DatoCMS migrations"

• Request "Find all mentions of image optimization across the docs"

NotebookLM excels at understanding relationships across documentation. Perfect for onboarding new team members or exploring features you haven't used yet.

Cursor & Windsurf: Code with full docs context

AI coding assistants need context about the tools you're using. That's where llms-full.txt becomes essential.

In Cursor:

1. Open Cursor and type @Docs

2. Select "Add new doc"

3. Paste: https://www.datocms.com/docs/llms-full.txt

In Windsurf:

1. Open settings → Documentation

2. Add new documentation source

3. Enter the URL: https://www.datocms.com/docs/llms-full.txt

Critical tip: Always type the @ symbol manually in the chat interface. Copy-pasting breaks the context reference.

Now you can:

• Type "create a Next.js page that fetches blog posts from DatoCMS" and get working code

• Ask "add pagination to this query" and it knows the DatoCMS pagination API

• Debug GraphQL queries with full knowledge of available fields and filters

Your AI coding assistant now knows DatoCMS as well as you do. Maybe better.

What makes it work

You might be thinking: "Can't I just scrape the docs?" Sure. But you'll get broken formatting, navigation menus mixed with content, and JavaScript cluttering everything.

llms-full.txt is different — clean Markdown with every code block formatted perfectly, logical structure, complete context across all 500+ pages, and it's regenerated automatically with every docs update. No noise, just the information that matters.

Try it yourself

Here's everything you need:

• The file: https://www.datocms.com/docs/llms-full.txt

• Standard llms.txt: https://www.datocms.com/docs/llms.txt (just the index)

• Any docs page as Markdown: Just append .md to any URL (e.g., https://www.datocms.com/docs/content-management-api.md)

Drop it into your favorite AI tool and see the difference. Build something with it. Break it. Tell us what you think.

Last week Claude Skills launched. Next week? Who knows what's coming. As AI tools evolve at breakneck speed, we'll keep making our documentation work better with them.