You're moving. You photographed each box with KontexVoX, dictating its contents and its destination room. "Box 14: coffee maker, filters, blue mugs, cookbook. Kitchen." Arriving at the new place, the boxes are stacked in the hallway. You no longer know what goes where. The idea: have the AI read the descriptions and propose a storage plan — which box goes in which room, and in what order to unpack.

Prerequisites

Before you start, you need:

1. Photos enriched with KontexVoX. One photo per box, with the number, contents and destination room in the EXIF Description field. See What is an EXIF file?.
2. An AI agent with file-system access: Claude Code, Codex CLI, or Gemini. Not a web chatbot.
3. The photos exported to your computer via AirDrop or the Files app (the channels that preserve EXIF — see How to share without losing the EXIF).
A storage plan is spatial reasoning: what goes where, in what order. A web chatbot doesn't read your 50 box photos. You need an agent that processes the batch and understands the categories.

What you dictated: an example

Here's what the EXIF of five box photos contains:

IMG_0031.jpg → "Box 12: sheets, pillows, duvet. Bedroom." IMG_0032.jpg → "Box 14: coffee maker, filters, mugs, cookbook. Kitchen." IMG_0033.jpg → "Box 18: towels, shower gel, toothbrush. Bathroom." IMG_0034.jpg → "Box 22: TV, remote, HDMI cable. Living room." IMG_0035.jpg → "Box 25: drill, wall plugs, level. Garage."

These descriptions, you dictated while closing the boxes. The AI agent will group by destination room, propose an unpacking order (essentials first), and generate a document you follow on arrival.

The prompt

Read all the JPEG images in this folder. For each image, extract the EXIF Description field (which contains the box number and its contents). From the full set of descriptions, propose a storage plan for the new home. Group the boxes by destination room (kitchen, bedroom, living room, bathroom, garage). Propose a prioritized unpacking order (essentials first: kitchen, bedroom, bathroom). Generate a Markdown document with the plan.

What you get

A storage plan in Markdown. For each room, the list of boxes to place there. Then a prioritized unpacking order: kitchen first (to eat and drink), then bedroom (to sleep), then bathroom (to wash). The rest can wait.

You follow the plan on arrival. Boxes go straight to the right room instead of piling up in the hallway. You unpack the essentials first. You know where the coffee maker is on the first morning.

Recommended tool: Claude Code

Claude Code is good at spatial reasoning. To propose a logical storage plan, it understands object categories and their natural destination in a home.

Frequently asked questions

Does the AI guess the room if I didn't say it in my description?

It works less well. If you say "Box 14: coffee maker, filters, mugs", the AI can infer "kitchen". If you say "Box 22: misc", it can't guess. Hence the importance of always ending your description with the destination room.

Can I include the new home's floor plan?

Yes. Add to the prompt: "The new home has the following rooms: kitchen, living room, 2 bedrooms, bathroom, garage, basement." The AI adapts its plan to the actual home instead of guessing.

Is the unpacking order customizable?

Yes. If you arrive at 8pm and the first instinct is to assemble the bed before the coffee maker, say so in the prompt: "Prioritize the bedroom first, then the kitchen." The AI adjusts.

Can I do the same thing without KontexVoX?

Yes — you can note each box's contents on paper and draw up the plan by hand. But 50 boxes means 50 descriptions to copy out. Voice changes everything: you dictate while closing the box, and the plan gets built afterwards. That's why I created KontexVoX: so the move is documented at packing time, not after.

Going further

Create a searchable box index Generate an optimized storage plan for your garage Optimize an enriched-photo session Moving page