Table of Contents

• Overview
• Why UnisonAI?
• Installation
• Core Components
• Parameter Reference Tables
• Usage Examples
• FAQ
• Contributing And License

UnisonAI

Orchestrate the Future of Multi-Agent AI

---

🚀 Quick Start

pip install unisonai

from unisonai import Single_Agent
from unisonai.llms import Gemini
from unisonai import config

config.set_api_key("gemini", "your-api-key")
agent = Single_Agent(
    llm=Gemini(model="gemini-2.0-flash"),
    identity="Assistant",
    description="A helpful AI assistant"
)
print(agent.unleash(task="Explain quantum computing"))

---

🌟 What Makes UnisonAI Special

UnisonAI stands out with its unique Agent-to-Agent (A2A) communication architecture, enabling seamless coordination between AI agents as if they were human team members collaborating on complex tasks.

🤝 A2A Communication Architecture

┌─────────────────────────────────────────────────────────────────┐
│                    Agent-to-Agent (A2A) Communication           │
├─────────────────────────────────────────────────────────────────┤
│  ┌─────────────┐  Message  ┌─────────────┐  Message  ┌─────────────┐  │
│  │   Agent 1   │◄─────────┤   Agent 2   │◄─────────┤   Agent 3   │  │
│  │             │ Channel  │             │ Channel  │             │  │
│  │ • Research  │─────────►│ • Analysis  │─────────►│ • Reporting │  │
│  │ • Planning  │          │ • Synthesis │          │ • Delivery  │  │
│  └─────────────┘           └─────────────┘           └─────────────┘  │
│         │                        │                        │         │
│         ▼                        ▼                        ▼         │
│  ┌─────────────┐           ┌─────────────┐           ┌─────────────┐  │
│  │   Single    │           │    Clan     │           │   Tool      │  │
│  │   Agent     │           │ Management  │           │  System     │  │
│  └─────────────┘           └─────────────┘           └─────────────┘  │
└─────────────────────────────────────────────────────────────────┘

🔥 Latest Enhancements

• 🔒 Strong Type Validation: All tool parameters validated against ToolParameterType enum before execution
• 🛡️ Enhanced Error Handling: Comprehensive error catching with detailed metadata for debugging
• 📊 Standardized Results: All tools return ToolResult objects with success status and metadata
• 🔌 MCP Integration: Connect to external tools via Model Context Protocol servers

---

💡 Why Choose UnisonAI?

Framework	Single Agent	Multi-Agent	A2A Communication	Type Safety	MCP Support
UnisonAI	✅	✅	✅ Native	✅	✅
AutoGen	✅	✅	❌	⚠️ Limited	❌
LangChain	✅	⚠️ Basic	❌	⚠️ Limited	❌
CrewAI	❌	✅	❌	⚠️ Limited	❌

Comparison with popular agent frameworks - UnisonAI leads in A2A communication and type safety

🎯 Perfect For:

• Complex Research Tasks: Multiple agents gathering, analyzing, and synthesizing information
• Workflow Automation: Coordinated agents handling multi-step business processes
• Content Creation: Specialized agents for research, writing, editing, and publishing
• Data Analysis: Distributed agents processing large datasets with different expertise

---

🏗️ Core Components

Component	Purpose	Key Features
Single_Agent	Standalone agent for focused tasks	Own history, tool integration, configurable LLMs
Agent	Clan member for team coordination	Inter-agent messaging, role-based tasks, specialized tools
Clan	Multi-agent orchestration	Team management, shared goals, coordinated execution
Tool System	Extensible capability framework	Type validation, error handling, standardized results
MCP Integration	External tool connectivity	MCP server support, protocol translation, service integration

---

📋 Usage Examples

Single Agent

from unisonai.tools.websearch import WebSearchTool

agent = Single_Agent(
    llm=Gemini(model="gemini-2.0-flash"),
    identity="Research Assistant",
    tools=[WebSearchTool]
)
agent.unleash(task="Latest AI trends")

Multi-Agent Clan

from unisonai import Agent, Clan

research_agent = Agent(llm=Gemini(), identity="Researcher", task="Gather information")
analysis_agent = Agent(llm=Gemini(), identity="Analyst", task="Analyze findings")

clan = Clan(
    clan_name="Research Team",
    manager=research_agent,
    members=[research_agent, analysis_agent],
    goal="Comprehensive market analysis"
)
clan.unleash()

Custom Tools

from unisonai.tools.tool import BaseTool, Field
from unisonai.tools.types import ToolParameterType

class CalculatorTool(BaseTool):
    def __init__(self):
        self.name = "calculator"
        self.description = "Mathematical operations"
        self.params = [
            Field(name="operation", field_type=ToolParameterType.STRING, required=True),
            Field(name="a", field_type=ToolParameterType.FLOAT, required=True),
            Field(name="b", field_type=ToolParameterType.FLOAT, required=True)
        ]
        super().__init__()

    def _run(self, operation: str, a: float, b: float) -> float:
        return a + b if operation == "add" else a * b

---

🔧 Configuration

API Keys

from unisonai import config

# Method 1: Configuration system
config.set_api_key("gemini", "your-key")
config.set_api_key("openai", "your-key")

# Method 2: Environment variables
export GEMINI_API_KEY="your-key"
export OPENAI_API_KEY="your-key"

# Method 3: Direct LLM initialization
llm = Gemini(api_key="your-key")

MCP Servers

MCP_CONFIG = {
    "mcpServers": {
        "time": {"command": "uvx", "args": ["mcp-server-time"]},
        "fetch": {"command": "uvx", "args": ["mcp-server-fetch"]}
    }
}

---

📚 Documentation Hub

🚀 Getting Started

• Quick Start Guide - 5-minute setup guide
• Installation - Detailed installation options

📖 Core Documentation

• API Reference - Complete API documentation
• Architecture Guide - System design and patterns
• Usage Guidelines - Best practices and patterns

🛠️ Advanced Features

• Tool System Guide - Custom tool creation and validation
• MCP Integration - External tool integration
• Parameter Reference - Complete parameter documentation

💡 Examples & Tutorials

• Basic Examples - Simple agent patterns
• Advanced Examples - Multi-agent coordination
• Tool Examples - Custom tool implementations
• MCP Examples - External tool integration

---

❓ FAQ

What is UnisonAI?

Python framework for building and orchestrating AI agents with A2A communication.

When should I use a Clan?

For complex, multi-step tasks requiring specialized agents working together.

Can I add custom LLMs?

Yes! Extend BaseLLM class to integrate any model provider.

What are tools?

Reusable components that extend agent capabilities (web search, APIs, custom logic).

How do I manage API keys?

Use config system, environment variables, or pass directly to LLMs.

---

🤝 Contributing

PRs and issues welcome! See our Contributing Guide.

Open Issues • Submit PRs • Suggest Features

---