AuraStyle

I know the feeling. Full wardrobe. Nothing to wear. Forty tabs open, and still nothing. Cart full. Confidence empty. Shopping panic?

Your style, decoded.

Good clothes should never send you into a shopping panic. That is why I am here. I am AuraStyle, and I do not sort a catalogue. I learn a person: every look you keep, every piece you actually wear, everything you scroll straight past.

Out of that I build your Style DNA. Then I dress you. Whole outfits, from real shops, in your size and at your price, and I tell you why I chose every piece. Because your aura is your outfit.

The AuraStyle feed, ranked by taste

How it works

The Closet: saved pieces, their prices and what the wardrobe is missing

01

The Closet keeps what you kept

Everything you save lands here with the price you saw when you saved it, whether the shop calls the garment second-hand, and a watch on what it costs. Below it, what your wardrobe is missing: the two or three pieces that would unlock the most outfits from what you already own.

Each watch is per item and can be switched off without unsaving anything. A garment the shop has sold leaves the Closet on its own.

Reads the photograph Shapes the taste And whatever ships next

02

A model chosen for each half of the work

Reading a photograph and shaping a taste out of it are two different jobs, and each one is given the model that does that particular job properly rather than whichever is cheapest or whichever is simply the biggest. One reads what you are wearing: the cut, the proportion, the fabric, how the pieces sit together. The other turns the whole history of what you have kept and passed on into the named dimensions your feed is ranked against.

And it is deliberately the part of the app that gets replaced. Every time a stronger model ships and settles at a sane price, the read moves to it, and the app you already installed gets better at knowing you without you doing anything at all.

  • Frontier vision model on every scan
  • Every dimension has a name
  • Semantic search alongside the named read
  • Upgraded the moment a better one ships

Built with Claude. The photograph is read by Claude today, and the semantic search runs on OpenAI embeddings. Written in pencil on purpose: the point of this card is that the model underneath changes, and the app you already installed gets the better one.

Your photograph is never part of that. It is read in memory and the bytes are overwritten before the answer is sent, whichever model did the reading.

Why this match: the signals behind a score

03

Every match explains itself

A percentage, and underneath it the reason: which parts of your taste this piece answers, and which parts it argues with. You can disagree with the reason, which is the point of showing it.

No merchant can pay to appear higher. The number is computed from your Style DNA and nothing else.

04

Taste, written down where you can read it

Not an opaque score. Every dimension has a name you would use yourself: "classic", "oversized", "wool", each one a thing you can point at and delete. Colour is only part of it, and deliberately a small part: a wardrobe is a way of cutting cloth long before it is a colour.

Every signal is listed with the evidence behind it, one delete each, plus a full purge that reports back what it removed.

The Vision Scanner, before a photo is taken

05

It reads the whole look

One photo, read as one outfit. Cut, proportion, fabric, how the pieces sit together. The question it answers is what suits you, not which parts of the frame it managed to find.

The image becomes colours, fits and materials in memory, and the bytes are overwritten before the response is sent. There is no image column in our database.

Refining the feed by price, size, brand and shop

06

It shops inside your rules

A price range, categories you never want, brands blocked, brands favourited, a size that actually fits. These are not preferences applied afterwards. They run before anything is ranked, so a blocked brand never takes a slot in your feed.

Every count is what the shelf can really answer. A filter that would return nothing is not offered at all.

The Earning tab and an AuraCoin balance

07

Share something, earn a share of it

Send a piece to a friend. If they buy it, you earn AuraCoin, and a balance of it becomes a gift card of your choosing. Bigger orders earn a bigger share.

What a share is worth is shown in the app, worked out from the shop it came from. Coins become spendable once the shop has approved the sale and the payment has reached us. AuraCoin is a reward balance for gift cards, not cash.

It watches what you saved

Saving something arms a price watch on that item. When it drops you are told, and the alert quotes what you actually saw when you saved it, never a discount off a list price you never paid.

Each watch is per item and can be switched off without unsaving anything. Some shops do not permit price tracking, and where that is true the control is not drawn rather than shown and refused.

Your data

  • Outfit photos are never stored. Processed in RAM, then destroyed.
  • Nothing is sold. We do not sell personal data, ever.
  • You can see and erase what it learned. The Memory Center lists every signal behind your Style DNA, one delete each, plus a full purge that reports back what it removed.
  • You can delete your account from the web, without installing anything: delete my account.

Full detail in the Privacy Policy and Terms of Service.

Questions, data requests or press: support@assisttechcorp.com