Pro as a whole doesn’t exist yet — but the MCP path already works today, for a wheel you’ve shared yourself with a code.

Let your AI assistant read and update a shared wheel

MCP (Model Context Protocol) is an open standard that lets an AI assistant — Claude, ChatGPT, Claude Code — read and edit data through a fixed address, a “server”. Kalindo has one. Share a wheel with a code in the editor, give the code to your assistant, and it can then see the rings and activities, add new ones, and give you an overview — with no access to anything beyond the one wheel the code points to.

How it works

1. Open the wheel and tap “Share via code” — you get a code like ABCD-EFGH-IJ.

2. Set up Kalindo’s MCP server in your assistant once (see below).

3. Ask your assistant to use the code, e.g.: “Use my Kalindo code ABCD-EFGH-IJ and tell me what’s happening in March.”

Set up the server

Claude — Settings → Connectors → add a custom connector, URL:

https://kalindo.dk/api/mcp

ChatGPT — Settings → Connectors → turn on developer mode → add connector, URL:

https://kalindo.dk/api/mcp

Claude Code — in your terminal:

claude mcp add --transport http kalindo https://kalindo.dk/api/mcp

Examples, once a wheel is shared

“What’s happening in March?”
“Add the workplace assessment in May to the Wellbeing ring”
“Give me an overview of the whole year”

What you should know

The wheel is stored on our server in the EU under a random code, so your assistant or colleagues with the code can read and change it. Don’t put sensitive personal information in the wheel. We delete it automatically 180 days after the last change.

You can stop sharing at any time with “Stop sharing” in the editor — the wheel then only stays in your browser, and the code stops working.