Welcome aboard. You've just installed an app that gets smarter every time you use it — so the sooner you start, the sooner it starts being genuinely useful.
#What It Does
ambientChat.ai is your personal AI context platform. That's a fancy way of saying: it remembers your stuff so your AI doesn't have to guess.
You can scan barcodes, photograph receipts, snap pictures of anything you own, and the app builds a rich, searchable catalog of your world. Then, when you ask a question — "What's expiring this week?", "How much did I spend at Costco?", "What spices do I have?" — your AI assistant already knows the answer, because it knows what you have.
The more you tell it, the better it gets. It's a bit like training a very eager, very organized personal assistant who never takes a day off and never forgets where you put the paprika.
#How to Use It
#iOS App
When you first open the app, you'll be asked to sign in. You can use Apple Sign-In, Google, or email. Pick whichever feels right — you can always change later from Settings.
Once you're in, you'll see a tab bar at the bottom of the screen with these sections:
#The Chat Tab
This is your AI assistant. Ask it anything about your inventory, your documents, or just general questions. The more context you've given it (by scanning things, uploading documents, being near beacons), the more useful its answers become.
The very first time you open Chat, there is no conversation yet — so the screen itself shows you what to try: a short greeting and three suggestion cards, with the message box below them. Nothing is covering anything, and there is nothing to close. Tap a card to go where it points, or ignore all of it and start typing — the cards are replaced by your first message either way.
There are seven suggestions in all and three on screen at a time. Tap the small ✕ on one you're not interested in and the next takes its place; the little row of dots underneath shows where you are in the set. The ones you dismiss stay dismissed, and if you change your mind, See all brings them back.
Once you've been using the app for a while, the same space becomes a short briefing instead — what's expiring, what's new, and a few questions worth asking. You can turn all of it off with Don't show again, and turn it back on from Settings → Application Behavior → Getting Started.
Try typing: "What do I have in my inventory?" — even if the answer is "nothing yet," you'll see it working.
#The Scan Tab
Point your camera at things and the app figures out what they are. Barcodes, receipts, products on a shelf, a pile of spices — Smart Scan handles all of it. This is the fastest way to build up your context.
#The Context Tab
This is where your data lives. It has four sub-tabs:
- Inventory — Everything you've scanned or added. A visual grid of your items with images, categories, and values.
- Documents — Receipts, manuals, warranties, and any files you've uploaded or the app has created. Organized in folders.
- Me — Your interests and preferences that help the AI understand you better.
- Questions — A history of questions you've asked and the answers you received. Your own personal Q&A archive.
#The Settings Tab
Your account, subscription, AI model preferences, privacy controls, data export, and all the knobs you might want to turn. You won't need this on day one, but it's good to know it's there.
#Your First Five Minutes
Here's what we'd suggest for your first session:
Scan something. Tap the Scan tab, point your camera at a product with a barcode, and tap the shutter button. Watch it appear in your inventory moments later.
Scan a receipt. Got a grocery receipt handy? Lay it flat, scan it, and watch the app extract the store name, individual items, prices, and date. Each item gets added to your inventory automatically.
Ask a question. Hop over to the Chat tab and try something like:
- "What's in my inventory?"
- "What did I just scan?"
- "What categories do my items fall into?"
Browse your stuff. Tap the Context tab, then Inventory. You should see your scanned items with AI-generated images and details. Tap any item to see its full details.
Explore Settings. Tap Settings to see your subscription tier, choose your preferred AI model, or adjust privacy settings.
That's it. Five minutes, and you're up and running.
#Via Chat
Your AI assistant is always ready. Here are some conversation starters once you have a few items:
- "What's expiring soon?"
- "How many items do I have?"
- "Show me my food items."
- "What did I buy at Target?"
- "Help me make a shopping list based on what I already have."
#Via Claude Desktop (MCP)
If you use Claude Desktop, you can connect ambientChat.ai as an MCP server. This lets Claude access your inventory, documents, and context directly from your desktop. It's the same data, the same intelligence — just a different window into it.
Once connected, you can ask Claude things like:
- "Search my inventory for electronics"
- "What documents do I have about my car?"
- "What food is expiring this week?"
See the MCP Tools guide for setup instructions.
#The app shows you around
The first time you open a screen, a few speech bubbles appear over its real buttons and say what each one does. Chat, Scan, and each part of your Context — Inventory, Docs, Q&A, Me and Trash — has its own, and Radar has one that names the things it finds around you.
Each bubble points at one control. Tap Next to move on, or the × to stop — it does not come back on its own, and stopping halfway is a perfectly normal thing to do.
On the web, the five parts of your Context have their own tours. They point at the web's own controls, so on a wide screen a bubble sits on the section in the sidebar, and on a phone-sized window it points at the menu button that holds them instead. Scan and Radar have no web tour — there is no camera or Bluetooth there to introduce.
Finishing a tour on your phone does not switch off the web one, and the other way round. The two are counted separately on purpose: the web puts things in different places, and knowing your way around the phone does not teach you that.
To see one again: Settings → Application Behavior → Getting Started → Show me around again on the phone, or Settings → App Behavior → Show me around again on the web. Either button re-arms every tour on both, and each runs the next time you open its screen. You can also just ask in chat — "show me around Chat again" — and the assistant will do it for you.
There is no way to switch them off in advance, and they need no switching off: each one runs once, on a screen you have not seen before, and never again unless you ask for it.
#Tips & Tricks
- The app learns from everything you scan. A receipt doesn't just create a document — it creates inventory items, tracks prices, and remembers where you shop. One scan, many benefits.
- Good lighting is your friend. Smart Scan works best in well-lit conditions. You don't need studio lighting, but scanning in a dark closet will test everyone's patience.
- On the web, rest the mouse on any icon to see what it does. After half a second a small label appears above round buttons like ••• , the trash can or the folder icon — the same words a screen reader says for that button. Buttons that already say what they do, like Upload, don't get one.
- You don't have to scan everything at once. Build your inventory gradually. Scan things as you buy them, and before you know it, your AI has a comprehensive picture of your world.
#Options
| Setting | What It Does | Default |
|---|---|---|
| AI Model | Choose which AI model answers your questions (OpenAI, Gemini, Claude) | Gemini Flash |
| Voice | Enable voice input and spoken responses | Off |
| Context Sharing | Control what context is shared with AI models | All enabled |
| Notifications | Expiry alerts, scan completion, beacon events | Enabled |
| Show me around again | Replay the guided tours — the speech bubbles that name the buttons on each screen. They run once on a screen you have not seen before; this is how you get them back. Settings → Application Behavior → Getting Started | — |
| Welcome suggestions | Show the greeting and suggestions at the top of a new chat — tips and what needs attention. Settings → Application Behavior → Getting Started on mobile, Settings → App Behavior on the web. Ticking "Don't show again" in the chat turns this off — this is how you turn it back on | Enabled |
#Using the app with VoiceOver
Every button in the iOS app now says what it is, and every box you can type in says which box it is.
That was not true until recently. Icon-only buttons — the zone chevrons on the Scan radar, the recentre button on the map, the ✕ on an alert card — announced only "button", and a text field announced "edit text, blank" whether it was the cost, the currency or the expiry date beside it. Nine switches in a row on a beacon's permissions all said "switch, on" and nothing about what was on.
A few things worth knowing:
- A field's name stays put once you start typing. The grey hint text in an empty box disappears the moment you type; the name VoiceOver reads does not.
- Switches say what they control, then their state — "Journey History, switch, on" rather than just "switch, on".
- The name you hear is the name printed on the screen. If a button is labelled "Open page", that is what VoiceOver says — not a different word for the same thing.
- VoiceOver says "button" after the name, so you can tell an action from a heading before you double-tap. It also means the rotor, Switch Control and Full Keyboard Access can jump between buttons — "next button" now reaches every control, where before it skipped most of them.
- Tabs say which one you are on — "Documents, button, selected" rather than five tabs that sound identical. The web app's tabs and its left-hand navigation do the same.
- Menus and dialogs open into their own contents. A ⋯ menu, a sort sheet, a picker, an alert card — swipe into one and you land on its first action, then swipe through the rest. Until this release a menu was announced as a single "Close menu" button with nothing inside it reachable, which meant no ⋯ menu anywhere in the app could be used without sight.
- Tapping outside a menu still closes it, and that backdrop announces itself as "Close menu" — so it is a deliberate way out rather than something you hit by accident.
On the iPhone, a card with its own buttons — an alert card for a tag you are near, a row with a share button, a thumbnail with a trash can — reads as the card and then its buttons, in order. Those buttons used to be inside the card and therefore unreachable. The card's own tap still works for everyone, and everything it did is also on a button you can reach: on a tag card, the link the card opens is the same link its Info or DPP button opens.
The web app has caught up. The same five places there — the share button on a skill row and on a folder row, Fix and Try again under a generated image, Renew on a share that is about to expire, and a collection card's or row's own move and trash buttons — sat inside the row or card they belong to, which a browser treats as one button and a screen reader cannot open into. Each is now its own button beside the row rather than inside it.
On the web a row is everything up to its own buttons: the title, the date, a folder name, and the arrow or ▶ at the end are all part of the row, so a tap on any of them opens the same thing. Only the buttons at the far right — share, move, trash, Renew — are separate, and each announces itself. The iPhone works the same way, with the whole row answering a tap; keyboard and screen-reader users reach every one of these either way.
An AI answer opens into its own buttons. On the iPhone a whole answer used to read as one thing: VoiceOver announced it, and Copy or select text, Listen, Save, Print, Export and Share — along with the record cards and sources underneath it — could not be reached at all. You could hear an answer and then have no way to keep it. Swipe through an answer now and you land on each of those in turn.
Two smaller things came with it. Copy or select text now appears on every answer, including ones where the app does not know which model replied — a restored conversation, or a note in the thread — so there is always a way to reach an answer's words. And your own messages can be copied without knowing the gesture: select one and activate it, and it copies. Holding it still works too.
A control that offers a choice now says which one you are on. Sort pills, filter chips, the Inventory/Documents toggle, "Can view / Can edit" in the share sheet, the model picker, the image-studio settings, the scan-rate and timeout segments — all of them highlighted the current option in colour and told a screen reader nothing, so you heard four identical buttons and had to remember which you picked. About a hundred of them now announce it: "Can edit, button, selected".
The web app had a second version of the same problem that looked fixed. Some of those controls carried the phone's way of saying it, which a browser silently throws away — so the code looked right, and a screen reader still heard nothing. Those say it in the web's own vocabulary now.
#Known Limitations
- The app requires an internet connection for scanning and AI chat. Offline mode is limited to browsing cached inventory.
- First-time sign-in with Apple occasionally requires a second attempt on some iOS versions.
- AI answers are only as good as the context you've provided — the more you scan and add, the more accurate and helpful responses become.
- Free tier users have a limited number of AI tokens per month. Upgrade to a paid tier for more.
#Version History
| Version | Date | What Changed |
|---|---|---|
| 12 | 2026-09-26 | Web: hover an icon to see what it does. Icon-only buttons across the web interface show a short label after half a second of hovering — the same name a screen reader announces (AMB-1273) |
| 10 | 2026-09-22 | Controls that offer a choice say which option is current. ~100 sort pills, filter chips, segment strips, the share sheet's permission and scope choices, and the model picker highlighted the active option visually and announced nothing — a screen reader heard N identical buttons. Five web controls looked fixed while using a property browsers discard, and now use the web's own |
| 9 | 2026-09-19 | An AI answer is no longer one VoiceOver element. Copy or select text, Listen, Save, Print, Export and Share sat inside the answer's own container, which iOS treats as a single element — so a blind user could hear an answer and reach none of the things you do with it. Copy or select text also now appears on answers whose model is unknown, and your own messages can be copied by activating them rather than only by holding them |
| 8 | 2026-09-18 | The web app's rows and cards open into their own buttons too. Share on a skill or folder row, Fix and Try again under a generated image, Renew on an expiring share, and a collection card's move and trash buttons were inside the row they belong to, which a browser treats as a single button — so a screen reader could not reach them |
| 7 | 2026-09-18 | Every menu, sheet, dialog and card on iOS now opens into its contents for VoiceOver. A ⋯ menu announced itself as one "Close menu" button and nothing in it could be reached — the same was true of the sort sheet, the save-to picker, the Contents list, the venue invitation and the tag alert card with its ten buttons. 18 screens fixed |
| 6 | 2026-09-16 | Every form field in the web app now says which field it is, instead of announcing "edit text, blank" and leaving you to guess — 114 of them, across the app and the admin console. A dozen had no placeholder either, so they said nothing at all |
| 5 | 2026-09-16 | VoiceOver now says "button" after a control's name on iOS, so you can tell an action from a heading, and "next button" on the rotor reaches every control instead of skipping most of them. Tabs also say which one you are on |
| 4 | 2026-09-13 | The iOS app is usable with VoiceOver. Every icon button now says what it is instead of just "button", every text field says which field it is instead of "edit text, blank", and every switch says what it controls before its state. Adds a "Using the app with VoiceOver" section |
| 3 | 2026-09-10 | The welcome suggestions are the first chat screen now, not a card on top of it. They used to arrive as a panel over the chat a few seconds after you opened it — which meant they landed after you had already started typing, and had to be dismissed before you could carry on. Now an empty conversation simply IS the greeting, three suggestions and the message box; your first message replaces them, so there is never anything to close. Three of seven show at a time, dismissing one brings the next in, and the chat background is white so the cards are actually legible |
| 2 | 2026-08-01 | Added the Welcome Card setting — the card's "Don't show again" checkbox previously had no way back |
| 1 | 2026-03-01 | Initial guide |