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

# Contextbase API Documentation

> Memory compression engine API for building context-aware LLM applications

## Welcome to Contextbase API

Contextbase is a **memory compression engine** that helps you build context-aware LLM applications. Publish unstructured data to our API and fetch data-rich, compressed context to prompt your LLMs with enhanced awareness and relevant information.

## Quick Start

Get started with the Contextbase API in just a few steps:

<Card title="Get Your API Key" icon="key" href="https://app.contextbase.dev/login">
  Sign up for a free account and get your API key to start making requests.
</Card>

## Core Features

Build powerful LLM applications with our memory compression engine:

<Columns cols={2}>
  <Card title="Resolve Contexts" icon="puzzle-piece" href="/api-reference/endpoint/resolve-context">
    Fetch compressed, context-rich prompts from your published data.
  </Card>

  <Card title="Publish Data" icon="upload" href="/api-reference/endpoint/publish-data">
    Publish unstructured data to contexts.
  </Card>

  <Card title="Scoped Context" icon="filter" href="/api-reference/introduction#core-concepts">
    Create personalized context using flexible scoping parameters.
  </Card>

  <Card title="RAG & Compression" icon="brain" href="/api-reference/introduction#preventing-context-rot">
    Prevent context rot with RAG filtering and agentic compression.
  </Card>
</Columns>

## Example Use Cases

See how teams are using Contextbase to enhance their LLM applications:

<Columns cols={2}>
  <Card title="AI Chatbots" icon="message-circle">
    Build chatbots with memory of user interactions, preferences, and conversation history.
  </Card>

  <Card title="Personalized AI Agents" icon="robot">
    Create AI agents that understand user context, project history, and relevant data.
  </Card>

  <Card title="Context-Aware Apps" icon="brain">
    Build applications that provide relevant information based on historical data and user context.
  </Card>

  <Card title="Dynamic Documentation" icon="book-open">
    Generate personalized documentation that adapts to user roles and current project state.
  </Card>
</Columns>

## How It Works

**Contexts** are data sources that contain information about specific domains or topics. **Prompts** combine multiple contexts in order to create comprehensive AI prompts:

1. **Publish** unstructured data to contexts (user messages, profile updates, system events, files)
2. **Scope** your data for personalization (by user\_id, project, environment, etc.)
3. **Resolve** individual contexts or entire prompts to get compressed, relevant content for your LLMs
4. **Query** with RAG to fetch only relevant information and help prevent context rot

## Need Help?

<Card title="Contact Support" icon="life-ring" href="mailto:support@contextbase.dev">
  Get in touch with our support team for assistance with your integration.
</Card>