Skip to main content

AI Integration Overview

Deep dive into LibreApps Desktop's AI architecture and capabilities.

Overview​

LibreApps Desktop is designed from the ground up to be an AI-first dashboard. This means that AI is not just an add-on, but a core part of the platform's architecture. This section provides a detailed look at how AI is integrated into LibreApps Desktop, from the backend services to the frontend components.

The AI Stack​

LibreApps Desktop's AI capabilities are powered by a robust and flexible stack:

  • LiteLLM: A unified interface for interacting with over 100+ AI models from various providers (OpenAI, Anthropic, Google, etc.).
  • AI Chat Server: A specialized backend service that handles chat history, streaming, and context management.
  • Smash AI Components: A set of pre-built React components for building AI-powered user interfaces.
  • Prompt Management: A centralized system for creating, testing, and versioning AI prompts.

Key Integration Points​

1. Conversational AI​

The most visible integration point, providing a real-time chat interface for users to interact with AI models.

2. Context Injection​

LibreApps Desktop automatically injects relevant page and user context into AI prompts, making the AI's responses more accurate and helpful.

3. Tool Calling (Function Calling)​

Allow AI models to interact with your application's APIs and data sources to perform actions on behalf of the user.

4. Background Processing​

Use AI for long-running tasks like document summarization, data analysis, and content generation.

Why This Architecture?​

  • Provider Agnostic: Easily switch between different AI providers without changing your code.
  • Scalable: Designed to handle high volumes of AI requests and concurrent users.
  • Secure: Keeps your API keys safe on the backend and provides granular control over AI access.
  • Developer Friendly: Provides a high-level API and pre-built components that make it easy to add AI to your dashboard.

Next Steps​

Explore the specific AI integration sections to learn more about how to configure and use each part of the stack.