Adding Items

Ohmventory has three ways to add an Item, all reachable from a Location's "Add" menu. Each one ends at the same place — an Item create form, pre-filled to different degrees — so you can mix and match depending on how many things you're adding and how much detail you want to enter by hand.

Manual add

The straightforward path: open a Location, choose "Add item," pick an Item Type, and fill in the form yourself — name, description, quantity, synonyms, owner, images, and whatever fields the chosen type defines. Good for one-off items or when you already know exactly what you're entering.

:::info Screenshot placeholder The manual item-create form with a type picker (searchable, showing the type hierarchy via indentation) and the resulting dynamic field list. :::

Add by AI (single photo)

Choose "Add by AI," take or upload one photo of the item, and Ohmventory analyzes it against your Organization's actual catalog of Item Types and Fields. It suggests:

• The best-matching Item Type (or up to three, if it's genuinely torn between plausible options).
• A suggested name, description, and quantity.
• Values for any type-specific fields it can confidently read from the photo (e.g. a resistor's color-band value, a label's printed part number).
• A generous list of synonyms, to make the item easier to find later via search.

You always review and can edit every suggested value before saving — nothing is created automatically without your confirmation, and the AI is instructed to leave a field blank rather than guess when it isn't confident. If none of your existing types are a good match, or the model returns no confident suggestion, you fall back to picking a type manually, with the photo already attached.

:::info Screenshot placeholder The AI suggestion review screen: the captured photo, a highlighted "AI suggested this" banner, and the pre-filled form fields below it, still editable. :::

Bulk add by AI

For logging many different items in one session at a Location — emptying a box of assorted parts onto a table and cataloging all of it, for example — "Bulk add by AI" lets you capture or select many photos up front (one photo per item, up to a soft cap), then works through the AI suggestions for each one, one at a time, in a review queue:

1. Capture phase: take photos one at a time, or select several existing photos at once from your device's library. Each photo is analyzed in the background as you keep capturing — you don't have to wait for one to finish before taking the next. A running count shows how many are queued, analyzing, ready, or failed.
2. Review phase: once you're done capturing, step through each photo's suggestion one by one — the same review form as the single-photo flow, with a "Save & Next" button that advances automatically, or "Discard" to drop that one entirely. Items whose analysis already finished are shown first, so you're never stuck waiting mid-review for a slow one.
3. Summary: once every item in the batch has been saved or discarded, a summary screen shows the final tally with links to everything you created, plus an option to start another batch right away for the same Location.

This mode is entirely session-based — if you close the tab or navigate away mid-batch with unreviewed items still pending, you'll be warned first, since progress isn't saved to the server until you actually save each individual item.

:::info Screenshot placeholder The bulk-add capture screen: a grid of photo thumbnails each with a status badge (queued / analyzing / ready / failed), and a "Done, review N items" button. :::

Which one to use

• One item, know the details already → manual add.
• One item, want help filling it in → Add by AI.
• Many different items at once (unpacking a delivery, cataloging a drawer's worth of parts) → Bulk add by AI.