Control Nanoleaf lights with ChatGPT and Claude via MCP

Someone using ChatGPT on a smartphone to set Nanoleaf lighting in a bedroom
Image: Nanoleaf.

Nanoleaf has announced an MCP server in an official press release that lets AI assistants such as ChatGPT and Claude control your bulbs, light panels and light strips in plain language. What separates it from a regular voice assistant is the freedom: you do not have to learn fixed commands, you can simply ask for a sunset feel in the living room. The assistant then works out which colours fit that description. There is one important caveat, though. Right now this is a job for developers, because the ready-made integrations are not due until October.

What works today, and who it is for

The MCP server and the developer setup guides are available from today at nanoleaf.me. So you set that server up by hand and then connect it to an assistant that supports MCP. Besides ChatGPT and Claude, Codex and the Gemini CLI are mentioned as well. That makes it interesting mainly if you have worked with an MCP server or a local API before. If you have not, waiting for the pre-built version is the sensible call.

Someone using ChatGPT on a smartphone to set Nanoleaf lighting in a bedroom
Image: Nanoleaf.

Lighting that reacts to weather and your calendar

The gain is not in switching a lamp on and off, because you do not need AI for that. The more interesting part is that an assistant can pull in outside data. Nanoleaf itself points to weather data, calendar entries and other web APIs as sources. If a storm rolls in, your lighting can shift to something warmer and calmer on its own, while a bright day calls for stronger light instead. You can hook in webhooks and third-party services too. That gets close to the kind of automation you would normally build in Home Assistant, except you now describe it in plain language rather than as a chain of conditions.

One-click integrations arrive in October

October 2026 brings the pre-built integrations for ChatGPT, Claude and other assistants. Nanoleaf describes those as being as easy to set up as an app extension, so no terminal and no server of your own. That is the point at which this becomes genuinely relevant for most people. It also continues a pattern you see across the category: Philips Hue picked up AI automations in early September, and Hue and Nanoleaf have been working together on Play Screen Sync since that same month.

Comments

No comments yet. Why don’t you start the discussion?

Leave a Reply

Your email address will not be published. Required fields are marked *