Capture your day. Kapel structures it.

Most tracking apps make you do the data entry. Kapel inverts that: you capture — a photo of a plate, a spoken set, a line of text — and the app composes the structured entry.

The idea

A meal isn't a calorie number; it's foods made of ingredients. A workout isn't a flat list of sets; it's zones with a protocol. Kapel models both honestly, so a photographed plate becomes ingredients with real macros, and an EDT block is tracked as the timed, rep-accumulating structure it actually is.

Built in the open, in alpha

Kapel is in alpha. Features described across this site are what ships today, tracked against a feature manifest that lives next to the code. Where a competitor does something better, the comparison pages say so — credibility is the point.

The MCP interface

Kapel exposes a Model Context Protocol server, so an AI assistant can read and write your training and nutrition data through the same typed tools the app uses — plan a week, log a session, or query your history in conversation.