You did a property inspection. Every room is photographed, every defect described out loud: "Kitchen, right wall, crack in the tiles above the counter, 10 cm." KontexVoX engraves the description into the EXIF, with date and time. Now you need to turn this set of photos into a professional PDF that landlord and tenant can sign.

An inspection without a structured document is photos in a phone. Nobody signs a phone. You need a document per room, with the defects listed, the reference photos, and signature fields at the bottom. Doing it by hand in Word is two hours of layout. With the descriptions already in the EXIF, the AI agent assembles the PDF in minutes.

Prerequisites

Before you start, you need:

1. Photos enriched with KontexVoX. The descriptions live in each file's EXIF Description field, with the capture date. 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 inspection PDF is layout: sections per room, defect table, signature fields. A web chatbot doesn't generate that kind of structured document from 40 photos. You need an agent working inside your folder.

What you dictated: an example

Here's what the EXIF of four inspection photos contains:

IMG_0211.jpg → "Kitchen, right wall. Crack in the tiles above the counter, 10 cm." — 06/15/2026 10:42 IMG_0215.jpg → "Bathroom, bathtub. Chipped enamel at the bottom, 3 visible chips." — 06/15/2026 10:48 IMG_0218.jpg → "Living room, ceiling. No visible defect. Clean paint." — 06/15/2026 10:55 IMG_0222.jpg → "Bedroom, floor. Scratched parquet near the window, 4 crossed scratches, 30 cm long." — 06/15/2026 11:03

These descriptions, you dictated while walking through the apartment, room by room. KontexVoX transcribed and timestamped them. The AI agent will group the photos by room, structure the defects into a table, and generate a PDF with one section per room.

The prompt

Read all the JPEG images in this folder. For each image, extract the EXIF Description field, the date, and the geolocation if available. Generate a structured property inspection PDF with: a cover page (date, address), one section per room (inferred from the descriptions), a defect table with photo thumbnail and description, and signature fields on the last page. The PDF must be professional, sober, and printable.

What the AI produces with this prompt

A structured inspection PDF:

1. Cover page: inspection date, property address.
2. One section per room: kitchen, living room, bedroom, bathroom. The rooms are inferred from your descriptions ("Kitchen, right wall" → Kitchen section).
3. Defect table: for each room, a table with the photo thumbnail, the description you dictated, and the timestamp.
4. Signature fields: on the last page, areas for landlord and tenant.

The PDF is printable, signable, and usable as supporting evidence. You can print it for paper signature, or send it as an email attachment (the channel that preserves the original photos' EXIF).

Recommended tool: Claude Code

Claude Code excels at generating structured documents in PDF format. For this kind of official document, the reliability of the output format is essential.

Frequently asked questions

Is the PDF legally valid?

A property inspection has no legally imposed format in most places. What matters is that it's dated, signed by both parties, and documents the condition of the property. The generated PDF contains those elements. For stronger evidential weight, the original photos with their EXIF metadata (date, time, description) remain the documentary proof.

Can I add the meters (water, electricity)?

Yes. Add to the prompt: "Include a Meters section with the readings mentioned in the descriptions." If you photographed the meters while dictating the readings ("Water meter: 1234 m³, electric meter: 5678 kWh"), the AI extracts them.

How many photos for a complete inspection?

30 to 50 photos for a standard apartment: every room, several angles, defects in close-up, the meters, the keys. With KontexVoX in Capture Mode, you describe while walking — 50 enriched photos take 15 minutes.

Can I do the same thing without KontexVoX?

Yes — you can take photos and fill in an inspection template by hand. But the moment you take the photo is the moment the defect is most visible. If you write the description later, you forget details. Voice engraves the context at the exact moment — that's why I created KontexVoX.

Going further

Automatic before/after comparison between two inspections Generate a structured appraisal report Create an insurance claim file Optimize an enriched-photo session Inspections page