# Quick Start

## Overview

ATLAS can be experienced in three ways:

1. **Public Dashboard** — Explore the live demo
2. **Beta Access** — Full features with your own data
3. **Self-Hosted** — Complete control on your infrastructure

***

## 1. Explore the Public Dashboard

The fastest way to understand ATLAS:

**Visit:** [dashboard.atlas-ai.org](https://dashboard.atlas-ai.org)

**What you can do:**

* Browse the knowledge graph visualization
* Search across indexed concepts
* See how extraction works
* Understand the data structure

**Limitations:**

* Read-only access
* Public knowledge only
* No personal data upload

***

## 2. Join the Beta

For full ATLAS functionality with your own knowledge:

### Step 1: Request Access

Visit [atlas-ai.org](https://atlas-ai.org) and join the waitlist.

We're onboarding users in batches to ensure quality support.

### Step 2: Initial Setup

Once approved, you'll receive:

* Access credentials
* Setup instructions
* Onboarding support

### Step 3: Connect Your Data

Start with your highest-value sources:

| Source            | Setup Time | Value                    |
| ----------------- | ---------- | ------------------------ |
| Twitter Bookmarks | 5 min      | High (years of curation) |
| Apple Notes       | 10 min     | Medium-High              |
| GitHub Stars      | 5 min      | Medium                   |

### Step 4: Run Extraction

ATLAS processes your content:

```
Content → LLM Analysis → Concepts + Insights + Actions
```

Initial extraction takes 10-30 minutes depending on volume.

### Step 5: Explore Your Graph

* Search across everything you've saved
* Discover connections between ideas
* See your knowledge visualized

***

## 3. Self-Hosted (Advanced)

For complete control:

### Requirements

| Component | Minimum                 |
| --------- | ----------------------- |
| Python    | 3.11+                   |
| Node.js   | 18+ (for dashboard)     |
| Storage   | 1GB+ (scales with data) |
| RAM       | 4GB+                    |

### Installation

```bash
# Clone repository (when available)
git clone https://github.com/atlas-ai/atlas.git
cd atlas

# Install Python package
pip install -e .

# Install dashboard dependencies
cd dashboard
pnpm install
```

### Configuration

```bash
# Create config
cp config.example.yaml config.yaml

# Edit with your settings
# - API keys for LLM (Claude/OpenAI)
# - Data source credentials
# - Storage paths
```

### Running

```bash
# Start API server
uvicorn atlas.api.main:app --port 8888

# Start dashboard (separate terminal)
cd dashboard && pnpm dev --port 3333

# Or use CLI
atlas dashboard  # Terminal dashboard
atlas search "query"  # Direct search
```

***

## First Steps After Setup

### 1. Import Your First Source

Start with Twitter bookmarks (highest signal-to-noise):

```bash
atlas module import x_alpha --file bookmarks.json
```

### 2. Run Extraction

Process imported content:

```bash
atlas index
```

### 3. Explore

Search your knowledge:

```bash
atlas search "topic you care about"
```

### 4. Review Concepts

See what was extracted:

```bash
atlas concepts --limit 20
```

### 5. Check Actions

View extracted action items:

```bash
atlas actions --status pending
```

***

## Next Steps

Once you're comfortable with basics:

| Goal               | Next Step                         |
| ------------------ | --------------------------------- |
| Add more sources   | Connect Apple Notes, GitHub, etc. |
| Explore graph      | Use web dashboard visualization   |
| Monetize knowledge | Configure x402 protocol           |
| Automate           | Set up agent workflows            |

***

## Getting Help

### Documentation

You're reading it! Browse the sidebar for specific topics.

### Support

* Beta users: Direct support channel
* General: <support@atlas-ai.org>

### Community

* Twitter: [@atlas\_ai](https://twitter.com/atlas_ai)
* Discord: Coming soon

***

## Common First Questions

**Q: How long does initial processing take?** A: 10-30 minutes for typical bookmark collections (1-5K items)

**Q: Can I export my data?** A: Yes, always. SQLite database is portable.

**Q: Is my data private?** A: Yes. Local-first. Cloud features are opt-in.

**Q: What LLM does extraction use?** A: Claude by default. OpenAI supported.


---

# 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.atlas-ai.org/getting-started/quick-start.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.