As part of a succession, you need to list and describe a deceased person's belongings. Furniture, appliances, artwork, vehicles, keepsakes. You're in the house, you photograph each item and describe it out loud: what it is, where it came from, why it's there. KontexVoX engraves each description into the EXIF. Now you need to turn this set of photos into a structured estate inventory for the notary.

An estate inventory is a formal document: list of assets, description, reference photo, category, estimated value. The notary needs it for the division, for inheritance tax, for valuing the estate. Doing it by hand, item by item, is long and emotionally heavy work. But if every photo already carries a description, the AI agent can assemble the document.

Prerequisites

Before you start, you need:

1. Photos enriched with KontexVoX. The descriptions live in each file's EXIF Description field. Ideally: the item's name, origin, context ("solid oak sideboard, 1950s, inherited from grandma, small chip on the left handle"). 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).
An estate inventory is rigor and traceability. The notary needs reference photos with their descriptions. A web chatbot erases the EXIF on upload. You need an agent that reads the files locally.

What you dictated: an example

Here's what the EXIF of four photos taken during a house inventory contains:

IMG_0031.jpg → "Solid oak sideboard, 1950s, two doors, one interior shelf. Inherited from grandma. Small chip on the left handle. Living room." IMG_0042.jpg → "Limoges porcelain set, 12 plates + serving dish, blue floral pattern. Dining room sideboard." IMG_0058.jpg → "Oil on canvas, Brittany landscape, signed R. Le Goff 1978. Framed. Above the fireplace." IMG_0073.jpg → "Vintage Peugeot bicycle, 1970s, steel frame, tires need replacing. Basement."

These descriptions, you dictated while walking through the house, room by room. KontexVoX transcribed and engraved them into the file. The AI agent will extract the structured information, group by category (furniture, tableware, artwork, vehicles), and produce the document the notary expects.

The prompt

Read all the JPEG images in this folder. For each image, extract the EXIF Description field and the date. Generate an estate inventory in Markdown structured as follows: 1) Header (name of the deceased/beneficiary, inventory date), 2) Asset table with columns: Item, Detailed description, Photo (reference), Estimated value (empty column to fill in), 3) Grouping by category (furniture, appliances, artwork, vehicles, other), 4) Summary at the end of the document. The format must meet a notary's requirements.

What the AI produces with this prompt

An estate inventory in Markdown, structured in four sections:

1. Header: name of the deceased or beneficiary, inventory date.
2. Asset table: Item, Detailed description, photo reference, Estimated value (left empty — the appraiser or notary fills it in).
3. Grouping by category: furniture, appliances, artwork, vehicles, other. The AI infers the categories from your descriptions.
4. Summary: total number of assets, breakdown by category.

The document can go to the notary as-is, or export to PDF. Estimated values are left empty — the notary or an auctioneer fills them in after appraisal. The reference photos, with their EXIF descriptions intact, remain proof of the assets' condition at the time of the inventory.

Recommended tool: Claude Code

For a legal document requiring precision and rigor, Claude Code is the safest choice. The structure of an estate inventory must be impeccable.

Frequently asked questions

Does the notary accept an inventory in Markdown?

Markdown reads as plain text and exports to PDF in one click. Most notaries accept a well-structured PDF. If your notary expects a specific format (Word, Excel), ask the AI agent to generate that format directly by adjusting the prompt.

Do I have to estimate the value of the assets?

No, and it's better not to. The inventory documents what exists. Valuation is the notary's or auctioneer's role. Your descriptions provide the material for the appraisal (era, condition, origin), but the figures are theirs. The "Estimated value" column is left empty in the generated document.

Do the photos serve as evidence?

The photos with their EXIF metadata (date, place, description) document the assets' condition at the time of the inventory. They don't replace a bailiff's report, but they constitute a timestamped, descriptive record that can be attached to the file. If an asset's condition is disputed, the enriched photo is stronger than a photo without context.

Can I do the same thing without KontexVoX?

Yes — you can photograph the assets and write the inventory by hand. But in the context of a succession, you're often in a house that needs emptying, under emotional pressure, with little time. Voice lets you describe each item on the spot, while looking at it, without sitting down at a desk to write. That's why I created KontexVoX: document at the moment you see, not after.