At Scale

Batch-extract prompts from a folder of images

Reading one PNG at a time does not scale. Batch extraction reads the prompt chunks across a whole folder and turns them into a searchable index.

Try it now: AI Metadata Inspector

Free, no login — upload an image and read its embedded metadata in your browser.

What to know

For a handful of images, a single-file metadata reader is enough. Across thousands of generations it is not: opening each PNG by hand to copy its prompt is slow, and the results end up scattered in notes rather than searchable.

Batch extraction reads the embedded prompt and settings from every file in a folder in one pass, then stores them together so you can search by prompt fragment, model, or seed. It only works on files that still hold their metadata — stripped or converted copies contribute nothing.

How to recover it

  1. 1

    Gather the original files

    Batch extraction reads embedded metadata, so point it at your original PNG exports, not converted or downloaded copies.

  2. 2

    Spot-check one file first

    Confirm the prompt is present by reading a single image in the AI Metadata Inspector before processing the whole folder.

  3. 3

    Import the folder to extract in bulk

    Use a tool that parses prompt chunks across a folder on import, so every image’s prompt and settings land in one searchable place.

This is exactly what an AI-native catalog does: import a folder and every embedded prompt, model, and seed becomes a searchable field — recovery at library scale, not one file at a time.

Frequently asked questions

Frequently Asked Questions

Quick answers to the most common questions.

Related recovery guides

Never lose a prompt again

Numonic extracts the prompt, model, seed, and settings from every AI image on import and keeps them searchable — so recovery stops being a rescue mission and the prompt is always attached to the work.