Haive AI Agent Framework¶
🤖 Haive AI Agent Framework
Build intelligent AI agents with state-of-the-art tools, seamless integrations, and powerful orchestration patterns.
🚀 Quick Start¶
# Install the framework
pip install haive-core haive-agents
# Or install all packages
pip install haive[all]
from haive.agents.simple import SimpleAgent
from haive.core.engine import AugLLMConfig
agent = SimpleAgent(
name="helper",
engine=AugLLMConfig(temperature=0.7)
)
result = agent.run("Hello, Haive!")
from haive.agents.react import ReactAgent
from haive.tools.web import web_search
agent = ReactAgent(
name="researcher",
tools=[web_search]
)
from haive.agents.multi import MultiAgent
workflow = MultiAgent([
planner_agent,
executor_agent,
reviewer_agent
], mode="sequential")
📦 Package Ecosystem¶
The Haive framework is organized into focused packages that work together seamlessly:
Foundation & Infrastructure
The core engine powering all Haive agents with state management, graph workflows, and tool orchestration.
Augmented LLM Engine
State Schema System
Graph Workflows
Tool Registry
Agent Implementations
Pre-built agent types for common patterns like reasoning, multi-agent coordination, and RAG workflows.
SimpleAgent & ReactAgent
MultiAgent Orchestration
RAG & Memory Agents
Planning & Research
Tool Integrations
Extensive library of tools for web scraping, APIs, databases, file operations, and external services.
Web & API Tools
Database Connectors
File Operations
Custom Tool Framework
Model Context Protocol
Integration with MCP servers for enhanced AI capabilities and external tool connections.
MCP Server Management
Protocol Integration
Tool Discovery
Context Sharing
Game Environments
Game-based environments for training and testing AI agents with reinforcement learning support.
Game Engine Integration
Agent Training
Environment Simulation
Performance Analytics
Data Processing
Stream processing and real-time data workflows for handling large-scale data operations.
Stream Processing
Real-time Workflows
Data Connectors
Pipeline Management
Ready Solutions
Pre-configured agents and workflows for common business use cases and industry applications.
Business Workflows
Industry Solutions
Template Agents
Quick Deployment
Hierarchical Agent Protocol
Advanced protocol for hierarchical agent communication and coordination in complex systems.
Agent Hierarchies
Communication Protocols
Coordination Patterns
Scalability Features
🎯 Use Cases & Examples¶
Build agents that can research topics, analyze data, and generate comprehensive reports.
View Examples →
Automate business processes with intelligent agents that handle workflows and decision-making.
View Examples →
Create tutoring and educational support agents with personalized learning capabilities.
View Examples →
🌐 Framework Architecture¶
graph TB subgraph "🤖 Application Layer" A1[Business Apps] A2[Research Tools] A3[Game Bots] end subgraph "🧠 Agent Layer" B1[haive-agents] B2[haive-prebuilt] end subgraph "🛠️ Capability Layer" C1[haive-tools] C2[haive-mcp] C3[haive-games] C4[haive-dataflow] end subgraph "⚡ Infrastructure Layer" D1[haive-core] D2[haive-hap] end A1 & A2 & A3 --> B1 & B2 B1 & B2 --> C1 & C2 & C3 & C4 C1 & C2 & C3 & C4 --> D1 & D2 style D1 fill:#8b5cf6,color:#ffffff style D2 fill:#8b5cf6,color:#ffffff
📚 Documentation & Resources¶
Getting Started Guide →
Complete tutorial from installation to building your first agents.
Browse API Docs →
Comprehensive API documentation for all packages.
View Examples →
Real-world examples and tutorials for common use cases.
🔗 Quick Links¶
GitHub Organization: haive-ai
Community Discord: Join Discussion
PyPI Packages: Browse on PyPI
Documentation Hub: All package documentation centralized here
🆘 Getting Help¶
📖 Documentation
Start with our comprehensive guides and API reference.
💬 Community
Join our Discord community for discussions and support.
🐛 Issues
Report bugs or request features on GitHub.
📧 Contact
Reach out to the core team for enterprise support.
Note
Framework Status: Haive is actively developed and continuously updated. All packages are synchronized and tested together for seamless integration.