# How to use OmniAgent?

1. **Go to OmniAgent** : From the main navigation or your app home screen, open OmniAgent.

<figure><img src="/files/8lTBKelhRxIOE18IWmcY" alt=""><figcaption></figcaption></figure>

2. **Choose your mode** : Select Thinking for quick results, Pro for more detailed work, or Ultra for maximum quality and reasoning power.

<figure><img src="/files/cx6Haf1qRNn5eBqBH8Aq" alt=""><figcaption></figcaption></figure>

3. **Describe what you want to create** Type your prompt in plain language — *"a 30-second ambient music track with a calm, cinematic feel"* or *"a photorealistic image of a futuristic city at dusk".*

<figure><img src="/files/ybDYJzLPuyQJrqgNXdf8" alt=""><figcaption></figcaption></figure>

4. **Upload a reference file (optional)** Attach an image, audio clip, or document to give OmniAgent a creative anchor to work from.

<figure><img src="/files/64mMiSSVMmUMzgYrkJit" alt=""><figcaption></figcaption></figure>

5. **Hit Generate** OmniAgent spins up your sandbox and begins working. Watch the agent's progress in real time as it plans and executes your request.

<figure><img src="/files/UyrCKLUvopDHeTiOUizW" alt=""><figcaption></figcaption></figure>

6. **Review the output** Your image, video, or music file will appear directly in the chat once complete.

7. **Refine in conversation** Describe any changes — *"make it warmer in tone"* or *"add more bass"* — and OmniAgent will iterate without starting over.

8. **Switch modes if needed** If your task grows more complex or you want a higher quality pass, upgrade to a higher mode and re-run.

9. **Download your output** Save your final file directly from the chat window.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://help.chatlyai.app/products-and-features/omniagent/how-to-use-omniagent.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.