AI Agents

LLMs.txt

Give AI tools a focused or complete text version of the Onderwijsregio API documentation.

What is LLMs.txt?

llms.txt is a convention for publishing documentation in a format that is easy for large language models to read.

Instead of asking an AI tool to discover every page through a website, you can point it at a curated text file containing the important concepts, guides, and links.

This is useful when an AI coding assistant needs reliable context about the Onderwijsregio API, data model, or permissions.

Available files

The documentation site publishes two files:

/llms.txt is the compact starting point. It contains the key guidance and links and is suitable for most tools.
/llms-full.txt contains the complete generated documentation, including detailed guides and API reference pages. It is substantially larger.

Start with /llms.txt. Use /llms-full.txt when the agent needs details from several parts of the documentation and its context window can handle the larger file.

Add the documentation to an AI tool

The exact workflow depends on the tool. In general, add the URL as documentation, web context, or a project reference, then ask the agent to use it when answering API questions.

Cursor and Windsurf

Use the tool’s documentation or web-context feature to add one of these URLs:

https://docs.onderwijsregio.onderwijsin.nl/llms.txt
https://docs.onderwijsregio.onderwijsin.nl/llms-full.txt

You can also mention the URL directly in a prompt when you want the context for a single task.

Other AI tools

For an assistant that accepts URLs as context, include a short instruction such as:

Use https://docs.onderwijsregio.onderwijsin.nl/llms.txt as the source of truth for Onderwijsregio API endpoints, models, and integration guidance.

If the tool supports MCP, consider connecting the Onderwijsregio API MCP server instead. MCP lets the agent discover only the reference entries it needs.

Keep the source of truth clear

The files are generated from the published documentation. They are a context aid, not an API credential and not a replacement for the live API response.

Tell your agent to verify important details against the linked reference page and to account for your token’s permissions. Some datasets and services require additional access.

When you ask an agent to write code, provide the intended use as well. For example, say whether you are retrieving articles, searching FAQs, building a programme finder, or working with media.

The compact file usually gives an agent enough orientation. Add individual documentation pages or use MCP when you need a precise endpoint, schema, or example.

MCP Server provides read-only, tool-based documentation discovery.
Getting started explains the available services and the first integration steps.

MCP Server

Give AI agents read-only access to the Onderwijsregio API documentation and reference.

Troubleshooting

Resolve common Onderwijsregio API integration issues, including rate limits, schema latency, record errors, and attachment uploads.