🚀 Welcome to Haive Agents Documentation¶

Haive Agents is the most advanced AI agent framework available today, providing truly dynamic agents capable of self-modification, self-replication, runtime adaptation, and autonomous coordination.

🧠 Build Intelligent Agents That Think, Learn, and Evolve

From simple conversations to complex multi-agent systems with graph-based memory

🔧 Self-Modifying 🤖 Self-Replicating 👥 Multi-Agent 🧠 Graph Memory

🎯 Quick Navigation¶

🚀 Quick Start

Get up and running with your first intelligent agent in minutes

Perfect for beginners - Step-by-step examples from basic chat to self-modifying agents

🚀 Quick Start Guide

💻 Installation

Complete setup guide for development and production

Production ready - Docker, monitoring, error handling, and scaling

💻 Installation & Setup

🎓 Examples & Tutorials

Comprehensive examples for every use case

Learn by doing - From simple agents to complex multi-agent systems

🎓 Examples & Tutorials

🧠 Agent Types

Complete guide to all available agent types

Choose your agent - Simple, ReAct, Multi-Agent, Self-Modifying, and more

Agent Types

📚 Memory Systems

Advanced memory and knowledge management

Intelligent memory - Graph-based knowledge with automatic relationship discovery

Memory Systems

👥 Dynamic Supervision

Self-organizing agent teams and coordination

Autonomous teams - Agents that build and manage other agents

Dynamic Supervisor System

Key Features¶

🔥 What Makes Haive Agents Special

True Intelligence: Not just chatbots - agents that reason, plan, and solve complex problems

Dynamic Adaptation: Agents modify their behavior at runtime based on performance

Self-Organization: Create agents that spawn and coordinate specialized teams

Advanced Memory: Graph-based knowledge systems with automatic relationship discovery

Production Ready: Comprehensive error handling, monitoring, and scaling capabilities

# Create your first intelligent agent in 3 lines
from haive.agents import SimpleAgent
from haive.core.engine.aug_llm import AugLLMConfig

agent = SimpleAgent(name="assistant", engine=AugLLMConfig())
result = await agent.arun("Explain quantum computing")
# Agent provides comprehensive, contextual response

User Guide¶

📖 Getting Started

🧠 Core Concepts

🔧 Advanced Topics

API Reference¶

🔍 API Reference

agents