# What is RAG? (And why it’s awesome)

Welcome to one of the brains behind Calk!\
Let’s break down **RAG** — in plain English.

***

### 🧩 RAG stands for "Retrieval-Augmented Generation"

It sounds fancy, but here’s the simple idea:

Imagine you’re asking ChatGPT a question. It tries to answer from what it already knows.

Now imagine if, before answering, it could **go look into your company’s Notion, Google Drive, or Slack**, find the exact info it needs — and **then answer** using that real information.

That’s RAG. 🧠🔍✍️

***

### 🧠 Why do you need RAG?

Here’s the thing:\
**AI models like GPT or Claude have a memory that doesn’t update.**\
They were trained months ago — so they don’t know:

* Your latest processes
* Your current clients
* What your team said in Slack last week

But your company moves fast.\
You write new docs, launch new features, and change how things work — sometimes every day.

**RAG lets your AI stay in sync.**

By connecting to your tools, it can:

* Retrieve your real, live knowledge
* Answer based on what’s true *right now*
* Keep learning without retraining the model

So your AI doesn’t guess — it **knows.**

***

### 🛠️ How does RAG work?

1. **You ask a question.**\
   → Like: “What’s our refund policy?”
2. **Calk searches your connected knowledge.**\
   → It looks through Notion pages, Google Docs, uploaded files, etc.
3. **It finds the most relevant info.**\
   → Like a sentence in your company handbook.
4. **Then it generates an answer using that info.**\
   → So it feels like it really understands your business.

That’s Retrieval-Augmented Generation:\
**It retrieves before it generates.**

***

### 🔗 How Calk uses RAG

Every time your agent answers with internal knowledge, RAG is doing the work in the background.

You don’t have to configure anything fancy:\
Just connect your tools → your agents become **faster, smarter, and always up to date.**

> Want to see it in action?\
> Check out the [Use Cases](/use-case.md) tab for real examples — from onboarding and customer support to async Q\&A and internal helpdesks.


---

# 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://docs.calk-ai.com/agents/what-is-rag-and-why-its-awesome.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.