AURA's features in full
AURA is a discussion companion in the shape of a 3D character that lives on your Windows desktop. Below is everything it can do, plainly — including the things it deliberately won't.
One thing to get straight first
AURA is someone to talk and think with — not an assistant that performs tasks. It does not control your computer, open applications, or modify files, and it won't pretend it can. Ask it to do something on your system and it will tell you it's a discussion companion only. That's a design decision, not a temporary limitation.
A discussion partner that pushes back
AURA's built-in character is made to help you think: to work through ideas, question assumptions, weigh trade-offs, and offer honest counterpoints — kindly, and without empty flattery. Replies are deliberately short and conversational unless you ask it to go deeper.
- It follows whatever language you write in — in Bahasa Indonesia/English; when that's unclear it replies in Bahasa Indonesia.
- This core character stays the same across every provider — only the model's reasoning quality changes.
A 3D character that genuinely lives on your desktop
AURA is not a static icon. It stands, walks around, and sits on its toolbox, with animations that follow its state. Leave it alone for a while and it gets bored and wanders off on its own.
- The boredom delay is adjustable — 60 seconds by default before it starts wandering.
- Walking speed is adjustable — 80 pixels per second by default.
- It moves across your desktop without covering the window you're working in.
- It can be hidden from Settings whenever you need a clear screen.
Bring your own API key (BYOK)
AURA doesn't resell AI quota. You supply your own API key, and conversations flow straight from your computer to that provider. There are five options, including one that runs entirely on your machine with no key at all.
- Anthropic (Claude), OpenAI, Google Gemini, OpenRouter, and local Ollama.
- The model name is a free text field, not a locked dropdown.
- You can switch providers at any time without reinstalling.
Chat with attachments, markdown, copy and save
AURA's chat bubble is built for real conversations, not one-liners. Long answers stay readable and easy to take with you.
- Attach up to 3 files to a single message to discuss them together.
- Replies render as markdown — lists, emphasis, and code blocks come out formatted.
- A copy button to grab an answer, and a save button to write it out as a file.
- While you're typing and discussing, AURA stops wandering so it doesn't distract you.
Memory you switch on yourself
By default AURA remembers nothing: every session starts fresh. If you'd like it to recall earlier conversations, turn that on in Settings — it's opt-in by design, for privacy.
- Off by default. Nothing is stored until you enable it.
- When enabled, history is kept locally on your device.
- There's a button to forget everything whenever you want.
Daily reminders
AURA can keep a simple list of reminders and raise a desktop notification when each one is due.
- Each reminder has a label and a time (24-hour format).
- One-off, or repeating on chosen days of the week.
- Each reminder can be switched off without deleting it.
Battery warning
On a laptop, AURA watches your battery and tells you when it's running low. You choose the threshold — 20 percent by default.
Two languages, and any name you like
AURA's interface is available in Indonesian and English. You can also rename it — that name appears in the chat bubble header and is the name it introduces itself with.
- Pet name is free text, up to 24 characters.
- The language setting only changes the interface; AURA still replies in whatever language you type in.
Privacy as the default, not an add-on
AURA has no camera, no microphone, and sends nothing to Layar Digital's servers.
- No webcam access anywhere in the app.
- No analytics or telemetry SDK of any kind.
- Your API key is stored encrypted using Windows' built-in encryption, tied to your Windows account.
- With Ollama, your conversations never leave the computer at all.
How AURA works
From installing the app to understanding where your messages actually go.
Four steps to get started
Install from the Microsoft Store
AURA is distributed through the Microsoft Store for Windows, so installation and updates are handled by the Store. No installer to download from anywhere else.
Pick an AI provider and add your key
Open Settings, choose one of the five providers, and paste your API key. If you pick Ollama this step is skipped entirely — just make sure Ollama is running on your computer.
Adjust the name, language and behaviour
Rename the pet, choose your interface language, and set how fast it walks and how long it idles before getting bored. Turn memory on if you want it to remember your conversations.
Start the conversation
Click AURA to open the chat bubble and start talking. Attach files if there's something you'd like to go through together.
What happens when you send a message
Your message is combined with AURA's core character and sent straight from your computer to the API of the provider you chose. The reply comes back to the app and renders in the chat bubble. There's no middleman: no Layar Digital server sits on that path. If memory is on, excerpts of earlier conversation are included so AURA can follow along — and those excerpts live locally on your device, not in our cloud.
Where your API key is stored
Your key is stored encrypted using Windows' built-in encryption (DPAPI), tied to your Windows account, so other accounts on the same computer can't read it. There is no plaintext fallback: if system encryption isn't available, AURA refuses to save the key rather than writing it in the clear. The key is decrypted only as needed when a request is sent, and is never handed to the interface layer.
What's in the Settings menu
| Section | What you can set |
|---|---|
| Pet Identity | Pet name (up to 24 characters), and whether the pet is shown on the desktop. |
| Discussion AI (BYOK) | Provider, model name, API key, and the Ollama URL when using a local model. |
| Behavior | Boredom delay before wandering (60 seconds by default) and walking speed (80 pixels per second by default). |
| Memory | Turn conversation memory on or off, plus a button to forget everything. |
| Reminders | Add labelled reminders with a time and weekly repetition. |
| Battery | Low-power warning threshold (20 percent by default). |
| About | The app version currently installed. |
Just installed and don't have an API key?
AURA is usable right away at no cost: out of the box it points at a locally running Ollama. Install Ollama, pull a model, and you can start a discussion with no key and no internet connection.
Choosing an AI provider
AURA is provider-agnostic by design. All five give you the same AURA character; what differs is reasoning quality, speed, cost, and how far your data travels from your computer.
At a glance
| Provider | Default model in AURA | API key needed? | Cost | Best for |
|---|---|---|---|---|
| Anthropic (Claude) | claude-sonnet-5 | Yes | Pay per use | Long discussions and deeper reasoning |
| OpenAI | gpt-4o-mini | Yes | Pay per use | An affordable all-rounder |
| Google Gemini | gemini-2.0-flash | Yes | Pay per use | Fast replies for everyday chat |
| OpenRouter | openrouter/auto | Yes | Pay per use | Trying many models with one key |
| Ollama | llama3.2 | No | Free | Full privacy and offline use |
How to choose
If you use AURA to sharpen ideas and want a partner that genuinely challenges you, a top-tier model like Claude is worth the cost. If AURA is more of a light companion through the day, a small model like gpt-4o-mini or gemini-2.0-flash is far cheaper and still feels good. If you're not sure which model you want, OpenRouter lets you switch around without creating a new account each time. And if your data must not leave your computer at all — or you simply want zero cost — Ollama is the answer.
Model names aren't locked
The model field in Settings is free text. The defaults in the table above are only a starting point — when a provider releases a new model, you can just type its name without waiting for an AURA update.
About cost
You pay the provider directly for what you use, billed per token, not AURA or Layar Digital. Because AURA deliberately keeps its replies short, casual conversation generally costs very little — but usage still depends on the model and how long your discussions run, so keep an eye on your provider's dashboard early on. With Ollama the cost is zero, since the model runs on your own device.
Running Ollama locally
Ollama runs language models directly on your computer. AURA reaches it at the local address 127.0.0.1 on port 11434, which you can change in Settings if you run it elsewhere. Because everything runs on your machine there's no API key, no bill, and not a single byte of conversation leaving the computer. The trade-off: answer quality depends on the model you pull and how powerful your hardware is.
API key setup guide
AURA runs on the AI model you choose. You only need to add an API key once — here's a step-by-step guide for each provider. No programming required.
Treat your API key like a password
Never share your API key with anyone, and never paste it into a public chat or a screenshot. Anyone holding your key can spend your AI quota in your name. AURA stores it encrypted on your own device.
Why do I need my own API key?
AURA uses a BYOK (Bring Your Own API Key) model: your conversations go straight to the provider you choose, never through Layar Digital's servers. That means privacy and cost stay in your hands — you pay the provider directly for what you use, or nothing at all with local Ollama.
Steps per provider
Steps below verified July 2026. Each provider's interface can change — if a button is named differently, look for an "API Keys" menu in their dashboard.
Anthropic (Claude)
- Go to platform.claude.com (Claude Console), create an account or log in.
- Open Account Settings → API Keys (platform.claude.com/settings/keys).
- Create a new key, choose its expiration, then copy it.
- Open AURA Settings, choose the Anthropic provider, paste the key.
- Make sure your account has credit/billing active.
OpenAI
- Go to platform.openai.com/api-keys and log in.
- Click "Create new secret key", then copy it — the key is shown only once, so store it somewhere safe.
- Paste it into AURA Settings and choose OpenAI.
- Add credit/billing to your account so you can keep using it.
Google Gemini
- Go to aistudio.google.com/apikey and sign in with your Google account.
- Click "Create API key".
- Copy the key.
- Paste it into AURA Settings and choose Gemini.
OpenRouter
- Go to openrouter.ai and create an account or log in.
- Open the Keys page (openrouter.ai/keys), create a new key, then copy it.
- Paste it into AURA Settings and choose OpenRouter.
- Pick a model from OpenRouter's list — one key gives access to hundreds of models.
Ollamano API key
- Install Ollama from ollama.com.
- Run the command: ollama pull llama3.2 (or any model you want).
- In AURA, choose the Ollama option; the API key field disappears on its own.
- Make sure the Ollama URL matches the port Ollama uses on your computer — the default is http://127.0.0.1:11434.
Common problems
"Invalid API Key"
The key is wrong, or a space came along at the start or end. Copy the full key again and retry.
"Quota/credit exhausted"
Add credit or enable billing in your provider's dashboard. A valid key is still rejected when the balance is empty.
"Model not found"
The model name is mistyped, or for Ollama the model hasn't been pulled yet. Check the correct name in the provider's docs, or run ollama pull first.
Ollama doesn't respond
Make sure the Ollama app is running, then match the URL in AURA Settings to the address and port Ollama is using.
The key won't save
AURA refuses to store a key when Windows' built-in encryption isn't available — deliberately, so the key is never written in the clear. Make sure you're signed in with a normal Windows account.
What does it cost?
You pay the provider directly for what you use (per token), not AURA or Layar Digital. With Ollama running locally, there's no API cost at all.