Support – CURIOUS AI

Using the Web Search feature

Web Search allows the assistant to fetch fresh information directly from the internet while it answers your question. To turn it on, click the "+" button next to the message input to open the tools menu, then enable "Web Search". Once activated, every question you send will be answered using real-time information from the web instead of only the model's built-in knowledge. This is especially useful when you need the latest news, current events, live data, sports results, stock movements, weather conditions, or any topic where up-to-date information matters. You can leave Web Search enabled by default or toggle it on only when needed — the assistant will indicate when it has consulted the web while preparing an answer.

Prompt Enhance — writing better prompts automatically

Prompt Enhance is a helper that automatically rewrites the prompt you type before it is sent to the AI. It adds missing context, structure, tone and clarity so the assistant understands your intent more accurately and returns a higher-quality answer. To use it, open the "+" menu next to the message box and toggle "Prompt Enhance" on. You can keep it enabled all the time — short, vague prompts will be expanded into clearer, well-formed instructions, while already-detailed prompts remain largely untouched. This feature is a good habit for beginners and power users alike, because it reduces back-and-forth and helps you reach the right answer in fewer messages.

Uploading files and images in a conversation

You can share files and images with the assistant directly inside any chat. Click the "+" button next to the message input and choose "File Upload", then select the file from your device. Supported inputs include documents, PDFs and images. Once uploaded, the file is attached to your next message and the assistant reads it together with your prompt — so you can ask questions about the content, request a summary, translate it, extract data from tables, or describe the contents of an image. For best results, keep each file focused on a single topic and write a clear instruction along with the upload (for example, "Summarise this PDF in five bullet points").

Creating images with AI

The assistant can generate images from a text description. Open the "+" menu and select "Create Image", then type a clear description of what you want to see — mention the subject, style, colours, background and mood. For example: "A calm mountain lake at sunrise, watercolour style, soft pastel colours". Once the image is generated, you can refine it by asking for changes in plain language, such as "Make the sky more orange" or "Add a small wooden boat on the water". Iterating this way is often the fastest path to the exact picture you had in mind. The clearer and more specific your description, the closer the result will be to your intent.

Starting a new chat and managing past conversations

Every question does not need to live in the same chat. Use the "New chat" button in the sidebar to start a fresh conversation whenever you switch topics — this keeps the assistant focused and prevents older context from bleeding into a new task. All your previous conversations appear in the sidebar under "Recent chats". Hover over any chat to rename it so you can find it later, pin the important ones to the top, or delete the ones you no longer need. Clicking a past chat re-opens it exactly where you left off, so you can continue the same discussion at any time.

Changing your profile name, password and theme

Your account settings are available from the account menu. Click your avatar at the bottom-left of the sidebar to open it. From here you can open Profile to update your display name and change your password, and switch the interface between Light and Dark theme using the theme toggle. Choosing a theme is a personal preference — Light works well in bright environments, while Dark is easier on the eyes at night or on OLED screens. All these changes take effect immediately and stay with your account across devices when you log in again.

Getting further help

If your question is not covered by the articles above, or you run into a bug or want to request a new feature, you can always reach out to us. We read every message and usually reply within 2 to 3 business days. When reporting an issue, please describe what you were trying to do, what happened instead, and — if possible — attach a screenshot; this helps us resolve your case much faster.