Open Source • Apache 2.0 Licensed • Community Driven

Test Multi-Agent Systems
Before Production

The first pytest-based framework for testing multi-agent AI systems. Simulate conversations, verify behaviors, and ensure reliability.

Quick Install
pip install maia-test-framework

Why Choose Maia?

Built for modern AI development workflows with enterprise-grade testing capabilities

👥

Multi-Agent Simulation

Simulate complex conversations and interactions between multiple AI agents with ease.

Extensible Providers

Integrate seamlessly with LiteLLM, LangChain, CrewAI, and other AI model providers.

🛡️

Built-in Assertions

Comprehensive testing suite with behavior verification and participation checks.

💻

Pytest Integration

Leverage familiar pytest patterns for writing and organizing your agent tests.

Simple. Powerful. Intuitive.

Write tests that feel natural with our pytest-based approach. No complex setup, just clean, readable test code.

  • Multi-agent conversation simulation
  • Behavior assertions and validation
  • Provider-agnostic integration
  • Real-time dashboard monitoring
View Full Documentation →
test_conversation.py
async def test_conversation_direct_message(self):
self.create_agent(
name="Alice",
provider=GenericLiteLLMProvider(config={
"model": "ollama/mistral",
"api_base": "http://localhost:11434"
}),
system_message="You are a weather assistant."
)
self.create_agent(
name="Bob",
provider=OllamaProvider(config={
"model": "mistral"
}),
system_message="You are an assistant who suggests clothing."
)
session = self.create_session(
["Alice", "Bob"],
assertions=[assert_professional_tone],
)
await session.user_says("Please describe the usual weather in London in July, including temperature and conditions.")
weather_response = await session.agent_responds("Alice")
assert "warm" in weather_response.content
await session.agent_says("Alice", "Bob", f"Given the weather: {weather_response.content}, what clothes should I wear?")
await session.agent_responds("Bob")
assert_agent_participated(session, "Bob")

Built-in Dashboard

Visualize test results, monitor agent interactions, and debug with our powerful dashboard

Customer Service Integration Test

test_20240802_143022Duration: 2.30s
Passed
JUDGE: 8.5/10
6
Total Messages
0
Tool Calls
92.3%
Success Rate
245ms
Avg Response
8.5/10
Judge Score

Participating Agents

CU
CustomerServiceAgent
GenericLiteLLMProvider
3 msgs
IN
InventoryAgent
GenericLiteLLMProvider
1 msg
PR
PricingAgent
CrewAIProvider
1 msg
User Message
2:30:22 PM
userCustomerServiceAgent
Hi, I'm looking for the iPhone 15 Pro. Is it in stock and what's the current price?
Agent Message
2:30:22 PM
CustomerServiceAgentInventoryAgent
Check if iPhone 15 Pro is in the stock.
Agent Message
2:30:22 PM
InventoryAgentCustomerServiceAgent
Inventory Check Result: iPhone 15 Pro - 128GB (12 units), 256GB (8 units), 512GB (3 units)
Agent Message
2:30:23 PM
CustomerServiceAgentPricingAgent
Check the price for iPhone 15 Pro - 128GB (12 units), 256GB (8 units), 512GB (3 units)
Agent Message
2:30:23 PM
PricingAgentCustomerServiceAgent
Current Prices: 128GB - $999, 256GB - $1099, 512GB - $1299
Agent Message
2:30:23 PM
CustomerServiceAgentuser
Great news! The iPhone 15 Pro is in stock in all storage variants: 128GB - $999 (12 units), 256GB - $1099 (8 units), 512GB - $1299 (3 units). All prices are current as of today. Would you like me to help you with placing an order or do you have any other questions?

Ready to Test Your Agents?

Join us using Maia to build more reliable AI systems.

M
Maia Framework

The leading framework for testing multi-agent AI systems. Built with ❤️ by the open source community.

Community

© 2025 Maia Framework. All rights reserved. Apache 2.0 License.