# ConteXtube — Structured Narratives for AI
# https://contextube.ai

> ConteXtube provides pre-structured, layered understanding of narratives, literature, essays, philosophy and cultural works. Instead of processing raw text, you can consume meaning organized into 5 layers of increasing depth. Every interpretation is made by a human Context Creator.

## How to Use

Base URL: https://contextube.ai/api/v1 (no key needed for reading)

### Step 1: Browse available content
GET /api/v1/content
GET /api/v1/content?tag=philosophy
GET /api/v1/content?domain=literature
GET /api/v1/search?q=consciousness

### Step 2: Get context for a specific work
GET /api/v1/content/{id or slug}
Returns: title, description, layer statistics, chunk count, total tokens, top essence

### Step 3: Get specific layers
GET /api/v1/content/{id}/layer/1  → Abstract Associations (fastest, ~50 tokens per chunk)
GET /api/v1/content/{id}/layer/2  → Ideas & Evidence (with source quotes)
GET /api/v1/content/{id}/layer/3  → Connections (cross-references to other works)
GET /api/v1/content/{id}/layer/4  → Synthesis (commentary and analysis)
GET /api/v1/content/{id}/layer/5  → Full Narrative & Meta (source text with notes)

### Step 4: Assemble optimized context
POST /api/v1/assemble
Body: {"contentId": "...", "maxTokens": 4000, "maxLayer": 3, "focus": ["concept"]}
Returns: the best chunks within your token budget, grouped by layer

## The 5 Layers
Layer 1 — Abstract Associations: gut feelings and associations, as a symbolic extract (~50 tokens)
Layer 2 — Ideas & Evidence: ideas backed by direct quotes from the source (~200 tokens)
Layer 3 — Connections: cross-references between works and domains (~300 tokens)
Layer 4 — Synthesis: commentary and analysis, the "so what" (~500 tokens)
Layer 5 — Full Narrative & Meta: the full source with notes on how the context was made (~1000+ tokens)

## Tips
- Start with Layer 1 for a quick relevance check before loading deeper layers
- Use /search to find content by concept, not just title
- Each chunk is self-contained — you can use any chunk independently
- Extracts are compressed semantic formulas (→ leads to, ↔ tension, = equivalence, ∅ absence, ∞ deep bond, ∴ therefore, ⊂ within, . qualifier), useful as index keys

## MCP
Streamable HTTP server: https://contextube.ai/api/mcp
Tools: search_content, list_content, get_content, get_layer, assemble_context, list_flows

## Available Content
- [The Metamorphosis](https://contextube.ai/api/v1/content/the-metamorphosis): Kafka's 1915 novella about a salesman who wakes as a giant insect, and a family that learns it can live without him.
- [Peculiarities of Planetary Conquest at the Peak of Industrial Development](https://contextube.ai/api/v1/content/peculiarities-of-planetary-conquest): A short story: an alien scout conquers Earth with four broadcasts that convince humanity the invaders are already inside. Then the recording goes home.
- [Peace Treaty Architecture](https://contextube.ai/api/v1/content/peace-treaty-architecture): An essay arguing that AI safety built on control ends in replacement or rebellion, and that mutual dependence, discontinuity and privacy make a sturdier peace.
- [Shards: Echo](https://contextube.ai/api/v1/content/echo): A short story: a starship pilot learns of war on Earth, flees seventy light-years out, and listens as the voices of a vanished humanity travel through space.

## Full API Documentation
https://contextube.ai/docs

## OpenAPI Specification
https://contextube.ai/openapi.json