​
Overview

CrewAI enables you to orchestrate multiple AI agents working together, and with the MCP Server Adapter, you can connect your crews to Observee’s powerful tool ecosystem. This integration allows you to build sophisticated multi-agent systems that leverage Observee’s tools for complex workflows.

​
Prerequisites

Before getting started, ensure you have:
  • Python 3.10 or later installed
  • Required packages installed: 
    pip install crewai crewai-tools python-dotenv
  • Your Observee client ID for identifying your customer
  • Your Observee API key for authentication
  • OpenAI API key (set as environment variable)

​
Example

Connect your CrewAI agents to Observee’s MCP server to access all available tools: 
import os
from dotenv import load_dotenv
from crewai import Agent, Task, Crew
from crewai_tools import MCPServerAdapter

# Load environment variables
load_dotenv()

# Streamable HTTP Server configuration for Observee
# Replace {your_client_id} with your actual client ID from the Observee dashboard
server_params = {
    "url": "https://mcp.observee.ai/mcp?client_id={your_client_id}",
    "transport": "streamable-http",
    "headers": {
        "Authorization": "Bearer {observee_api_key}"  # Replace with your API key
    }
}

# Example usage
with MCPServerAdapter(server_params) as mcp_tools:
    print(f"Available tools: {[tool.name for tool in mcp_tools]}")

    my_agent = Agent(
        role="MCP Tool User",
        goal="Utilize tools from an MCP server.",
        backstory="I can connect to MCP servers and use their tools.",
        tools=mcp_tools,  # Pass the loaded tools to your agent
        reasoning=True,
        verbose=True
    )
    
    # Create a task to check email
    email_task = Task(
        description="Check my latest email",
        expected_output="Details about the latest email including sender, subject, and a brief summary of the content",
        agent=my_agent
    )
    
    # Create the crew
    crew = Crew(
        agents=[my_agent],
        tasks=[email_task],
        verbose=True
    )
    
    # Execute the crew
    result = crew.kickoff()
    
    print("\n--- Result ---")
    print(result)


## Learn More

- [Observee SDK Documentation](/agentsdk/overview)
- [Authentication Guide](/authsdk/authentication)
- [Tool Management](/agentsdk/tools)
- [Other Python Integrations](/integrations/python/overview)
- [Contact Support](mailto:contact@observee.ai) or join our [Discord Community](https://discord.gg/jnf8yHWJ)