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

# Knowledge Unification

## The Core Challenge

Your knowledge lives in silos:

* Twitter bookmarks you can't search effectively
* Apple Notes buried in folders
* Browser bookmarks with broken links
* Notion databases you forgot existed
* ChatGPT conversations that vanished

**ATLAS unifies everything into a single, searchable knowledge graph.**

***

## Supported Data Sources

### Currently Available

| Source                  | What's Captured                           |
| ----------------------- | ----------------------------------------- |
| **Twitter/X Bookmarks** | Tweets, threads, linked articles, images  |
| **Apple Notes**         | Full note content, attachments, folders   |
| **GitHub Stars**        | Repository metadata, READMEs, topics      |
| **LLM Conversations**   | ChatGPT, Claude exports with full context |
| **Voice Memos**         | Transcribed audio with AI extraction      |
| **Readwise**            | Highlights, annotations, and notes        |
| **Obsidian Vaults**     | Markdown notes with full link structure   |

### Coming Soon

* Kindle Highlights
* Pocket/Instapaper
* Notion Exports
* YouTube Watch Later
* Podcast Notes
* Telegram Groups (Alpha agent integration)

***

## The 5-Stage Pipeline

Every data source goes through ATLAS's standardized ingestion pipeline:

### Stage 1: Acquire

**Fetch raw data from the source**

Connect to APIs, import files, or process exports. ATLAS handles authentication, rate limiting, and data retrieval automatically.

### Stage 2: Prepare

**Normalize into consistent format**

Different sources have different structures. A tweet looks nothing like an Apple Note. This stage standardizes everything into a common schema.

### Stage 3: Process

**Extract intelligence with AI**

This is where the magic happens. LLMs analyze each piece of content to extract:

* Key concepts
* Actionable insights
* Important quotes
* Mental frameworks
* Importance scores

### Stage 4: Parse

**Structure into graph nodes**

Extracted intelligence becomes nodes in your knowledge graph. Concepts link to content, insights link to concepts, everything connects.

### Stage 5: Render

**Ready for search and visualization**

Final processing for full-text search indexing, graph visualization, and API exposure.

***

## What Gets Unified

### Content Items

The original sources — tweets, notes, documents, conversations.

### Concepts

Key ideas extracted from your content. Each concept has:

* Name and description
* Importance score (0.0 - 1.0)
* Source references
* Related concepts
* Associated insights

### Insights

Actionable learnings and observations. Things you should remember and act upon.

### Actions

Tasks, research threads, ideas to explore. Your knowledge becomes a to-do list.

### Relationships

Connections between everything. Concept A relates to Concept B. Content X contains Concept Y.

***

## The Unified Experience

### Before ATLAS

```
"I know I read something about this..."
→ Search Twitter (nothing)
→ Search Notes (nothing)
→ Search bookmarks (nothing)
→ Google it again
→ Start from scratch
```

### After ATLAS

```
"I know I read something about this..."
→ Search ATLAS
→ Find original content + extracted concepts + related insights
→ See connections to other things you've saved
→ Build on existing knowledge
```

***

## Key Benefits

### 🔍 Universal Search

One search box. Everything you've ever saved. Full-text search with semantic understanding.

### 🧠 Automatic Organization

No manual tagging or filing. ATLAS understands your content and organizes it automatically.

### 🔗 Connection Discovery

See relationships between ideas from different sources. Your Twitter bookmark connects to your Apple Note connects to your ChatGPT conversation.

### 📈 Growing Intelligence

Every new piece of content enriches existing knowledge. Your graph gets smarter over time.

### 💾 Data Ownership

Your unified knowledge lives in a SQLite database you control. Export anytime. No vendor lock-in.

***

## Technical Foundation

ATLAS uses a hybrid storage architecture:

* **SQLite** — Portable, reliable, local-first database
* **FTS5** — Full-text search with relevance ranking
* **Graph Schema** — Nodes and edges for relationship traversal
* **Vector Embeddings** — Semantic similarity (coming soon)

All running locally. All under your control.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## 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/features/knowledge-unification.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.