A lightweight Python framework for orchestrating multi-agent AI systems with A2A communication and multi-LLM support, built on Agent, Clan, and Tool components.

UnisonAI#

Overview#

UnisonAI is a lightweight Python framework designed to streamline the development of single and multi-agent systems. Its core innovation lies in the Clan concept for orchestrating multiple Agents to collaborate, supporting team-like task distribution and Agent-to-Agent (A2A) communication.

Tagline: "Orchestrate the Future of Multi-Agent AI"

Author: Anant Sharma (E5Anant) Current Version: v1.2 (Beta)

Core Components#

Component	Purpose	Key Features
Agent	Independent agent or team member	Tool integration, history, inter-agent messaging
Clan	Multi-agent orchestration	Auto-planning, task distribution, A2A communication
Tool System	Extensible capability framework	@tool decorator, BaseTool class, type validation

Key Features#

1. A2A Communication Architecture#

Agents collaborate like team members on complex tasks
Message passing and coordination between agents

2. Multi-LLM Provider Support#

Native support: Google Gemini, OpenAI, Anthropic, Cohere, Groq, Mistral AI, xAI (Grok), Cerebras, with custom model integration via BaseLLM extension.

3. Tool System#

Decorator Mode (Recommended): Quick definition using @tool decorator
Class Mode (Complex Scenarios): Inherit from BaseTool class
Strong Type Validation: Based on ToolParameterType and Pydantic
Standardized Results: Unified ToolResult object return

Installation & Quick Start#

pip install unisonai

Single Agent Example#

from unisonai import Agent
from unisonai.llms import Gemini

agent = Agent(
    llm=Gemini(model="gemini-2.0-flash"),
    identity="Assistant",
    description="A helpful AI assistant",
)

print(agent.unleash(task="Explain quantum computing in 3 sentences"))

Multi-Agent Team (Clan) Example#

from unisonai import Agent, Clan
from unisonai.llms import Gemini

researcher = Agent(
    llm=Gemini(model="gemini-2.0-flash"),
    identity="Researcher",
    description="Gathers information on topics",
)

writer = Agent(
    llm=Gemini(model="gemini-2.0-flash"),
    identity="Writer",
    description="Writes clear reports from research",
)

clan = Clan(
    clan_name="Research Team",
    manager=researcher,
    members=[researcher, writer],
    shared_instruction="Researcher gathers info, Writer produces the report.",
    goal="Write a brief report on AI in healthcare",
    output_file="report.txt",
)

clan.unleash()

API Key Configuration#

# Method 1: Config system
from unisonai import config
config.set_api_key("gemini", "your-key")

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

# Method 3: Direct initialization
from unisonai.llms import Gemini
llm = Gemini(api_key="your-key")

Custom Tool Creation#

from unisonai.tools.tool import tool

@tool(name="calculator", description="Math operations")
def calculator(operation: str, a: float, b: float) -> str:
    if operation == "add":
        return str(a + b)
    elif operation == "multiply":
        return str(a * b)
    return "Unknown operation"

Use Cases#

Complex Research Tasks: Multi-agent information gathering, analysis, and synthesis
Workflow Automation: Coordinating multi-step business processes
Content Creation: Specialized agents for research, writing, and editing
Data Analysis: Distributed agents processing large datasets

Core Dependencies#

Pydantic (≥2.4.2) - Data validation
Rich, Colorama - Terminal UI
nest-asyncio - Async support
Various LLM SDKs (openai, anthropic, google-generativeai, etc.)

UnisonAI