> For the complete documentation index, see [llms.txt](https://docs.calk-ai.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.calk-ai.com/readme-1.md).

# How to get started with Calk AI ?

### How it works ?

Calk AI is built around a simple idea:

*Connect your data → Build agents → Work smarter with AI.*

Whether you're training a sales assistant or automating internal research, everything in Calk AI follows this flow.

Let’s break it down.

### 1️⃣ Tag Any Model <a href="#h_1029892ef6" id="h_1029892ef6"></a>

Every conversation starts with the right AI model.

With Calk AI, just type `@GPT-4.1`, `@Claude3.7`, `@Mistral3medium`, or any other available model to activate it instantly.

You don’t need separate accounts. All the best models are in one place.

<figure><img src="https://2442470273-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FBGWv9tHN4FLRiQu7ou5P%2Fuploads%2FuJtPDbAptpLmkS6WXfsv%2Ftag-calk.png?alt=media&amp;token=60956b42-c651-4de8-b1c5-8aee4fc7f2c8" alt=""><figcaption></figcaption></figure>

### 2️⃣ Connect Your Data <a href="#h_a1dec7257a" id="h_a1dec7257a"></a>

AI is useless without context.

That’s why Calk AI lets you plug in your business knowledge — like documents, pages, notes, and links — from tools like Google Drive and Notion.\
Your data becomes the foundation for every agent you build.

Ask your first question just tag @gpt, @claude3.7 and @deepseek and ask them anything you'd like!

<figure><img src="https://2442470273-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FBGWv9tHN4FLRiQu7ou5P%2Fuploads%2FgVi2gC8tscUHt3NYZxNO%2FGroup%201030004908%20(1).png?alt=media&amp;token=f2abbcc1-8daf-416d-8b6e-5c5e6ddfd967" alt=""><figcaption></figcaption></figure>

***

### 3. Build Your Agents <a href="#h_09fcdf9627" id="h_09fcdf9627"></a>

Once your data is connected, you can:

* Create agents trained on it
* Or pick a ready-made template to skip the prompting

Agents can specialize in anything: sales, support, marketing, research…\
You define what they do — and they’ll answer based on the knowledge you gave them.

`Don't forget to link some data to the agent to maximize results!`

<figure><img src="https://2442470273-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FBGWv9tHN4FLRiQu7ou5P%2Fuploads%2FkhfUuwXZK6YxZqs2cQQy%2FGroup%201030004691.png?alt=media&amp;token=e51444e2-0640-4d19-a27d-5565f7f3e5b9" alt=""><figcaption></figcaption></figure>

### 4. Use One, Then the Next <a href="#h_09fcdf9627" id="h_09fcdf9627"></a>

Calk AI is built to work like a chain.

You can use one agent to generate something, then pass the result to another.\
One task at a time, each handled by a focused AI — no confusion, no overload.

That’s how you go from raw data to clean output — step by step, agent by agent.

<figure><img src="https://2442470273-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FBGWv9tHN4FLRiQu7ou5P%2Fuploads%2FxI4haNDiGAePpElUvtbg%2FGroup%201030005037.png?alt=media&amp;token=7c3e0a3d-368c-4fc2-991c-df148ce31da6" alt="" width="563"><figcaption></figcaption></figure>

***

Calk isn’t just software—it’s the future of how businesses interact with AI. Start building your custom AI solutions today and see the difference it makes for your team.