Skip to content
Home
How to Use with AI

How to Use with AI

Point your AI coding assistant at Maptoolkit’s llms.txt and it can add a working vector basemap to your project without you looking anything up. This works with any AI coding assistant — Claude, ChatGPT, Cursor, GitHub Copilot. (The Enterprise License comes with an MCP Server.)

Example Prompt

Before writing any code, retrieve and read the FULL raw contents of https://www.maptoolkit.org/llms.txt — use whatever capability you have: your built-in web fetch/browse tool, or a shell command like curl/wget if you have terminal access.

Do not summarize the file, do not delegate reading it, and do not rely on prior knowledge about Maptoolkit — the file is the only source of truth. If you cannot retrieve the file, stop and tell me — do not guess or invent URLs, style names, or API details.

Task: create an HTML page with a free Maptoolkit.org vector basemap using MapLibre GL JS, following the file’s MapLibre example including the mandatory attribution.

Why This Works

llms.txt is a single plain-text reference built for AI assistants to read in one pass. It contains the style URL, the full endpoint list, the attribution requirements, ready-to-use code for every major mapping library, migration snippets from Mapbox and Google Maps, and the license and rate-limit rules — so the assistant does not have to guess or rely on outdated training data.

Attribution (the Maptoolkit logo and copyright line) is required on every map under the Community License. A good AI assistant will add it automatically once it reads llms.txt — check the result against Step 2 of the Quick Start.

Next Steps