A first implementatin of Smart Graph on a simple chat app

Simple Chat Workflow

SmartGraph

SmartGraph is a powerful, reactive Python framework for building scalable Large Language Model (LLM) applications.  It simplifies complex AI workflows with modular components, built-in state management, and seamless LLM integration.  Inspired by modern frontend frameworks, SmartGraph brings reactive programming to AI development, enabling developers to create efficient, maintainable, and visualizable LLM pipelines.

Introduction

Introduction to SmartGraph

Get up and running with SmartGraph in minutes

Quickstart Guide

This document provides a comprehensive guide for developers who want to contribute to the SmartGraph library. It covers setting up your development environment, coding conventions, testing procedures, and other relevant information.

Development

Learn about the fundamental building blocks of SmartGraph, including ReactiveSmartGraph, Pipeline, and ReactiveComponent.

Core Concepts

Learn how to define transitions and implement dynamic decision-making in SmartGraph workflows

Pipeline

Controlling Flow: pipeline and Branching Logic

Learn how SmartGraph manages state within ReactiveComponents using BehaviorSubjects

State

State Management

Learn how to use DuckMemoryToolkit and extend MemoryToolkit for persistent storage in SmartGraph applications

Memory

Memory Toolkits

Learn how to integrate AI assistants into your SmartGraph applications using CompletionComponent

AI Assistants

Learn how to create and customize your own components in SmartGraph to extend its functionality

Custom Components

Creating Custom Components

Learn how to handle errors effectively and debug SmartGraph applications

Error Handling & Debugging

Error Handling and Debugging

Learn how to effectively test your SmartGraph applications, components, and pipelines

Testing

Testing SmartGraph Applications

Learn how to create and use custom toolkits with CompletionComponent in SmartGraph

Examples