> ## Documentation Index
> Fetch the complete documentation index at: https://docs.ascn.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Knowledge

> Everything the agent knows about itself and about you — before the first message.

Knowledge is what the agent reads before answering. Who it is, how it behaves, what you prefer, what documents to use.

<img src="https://mintcdn.com/ascn-b377dab5/gJOzWPSTEzSbhz3B/images/en/09-knowledge-base.png?fit=max&auto=format&n=gJOzWPSTEzSbhz3B&q=85&s=311ef226a6e8865a08fbee96f67d99ed" alt="Brain → Knowledge" width="2531" height="1380" data-path="images/en/09-knowledge-base.png" />

## What's here

**Core** — three system files: IDENTITY.md, SOUL.md and USER.md. They define personality, values and everything the agent knows about you.

**Documents** — reference files: client lists, templates, instructions, policies. Add via **+ Add** or drag and drop.

***

### IDENTITY.md — who your agent is

A job description. Here you set the role, communication style, and what the agent does and doesn't do.

```markdown theme={null}
# Role
You are an assistant for email and meetings.

## Style
- Short answers: max 3-4 sentences unless asked for more
- When unsure — ask, don't guess
- Business tone, no fluff
```

### USER.md — information about you

Everything the agent should know about you: name, projects, preferences, timezone. The agent updates this file itself — but better to fill it in upfront.

```markdown theme={null}
# About the user
Name: [name]

## Preferences
- Short answers, no fluff
- Timezone: UTC+0
```

To make the agent remember something right now, say in chat:

```text theme={null}
Remember: I prefer answers as a list, no intros
```

### SOUL.md — personality and priorities

Fine-tuning. How the agent behaves in ambiguous situations.

### Documents

Reference docs the agent uses when answering. Upload via **Documents → + Add**.

> How knowledge differs from memory — see [Memory](/en/brain/memory).