Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.observee.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Overview

The OpenAI Python SDK can connect to Observee’s MCP (Model Context Protocol) server using the agents library. This enables your Python-based agents to access Observee’s powerful tool ecosystem, integrate with internal or external APIs, and leverage custom knowledge sources through a unified interface.

​
Prerequisites

Before getting started, ensure you have:
  • Python 3.10 or later installed
  • Required packages installed: 
    pip install openai agents python-dotenv
  • Your Observee client ID for identifying your customer
  • Your Observee API key for authentication
  • OpenAI API key

​
Example: Connecting an Agent to Observee’s MCP Server

Below is a complete example of how to set up an Agent that communicates with your Observee MCP server using Python: 
import asyncio
from agents import Agent, Runner, set_default_openai_client
from agents.mcp import MCPServerStreamableHttp
from openai import AsyncOpenAI
from openai.types.responses import ResponseTextDeltaEvent
import os
from dotenv import load_dotenv

load_dotenv()

async def main():
    # Set up OpenAI client
    client = AsyncOpenAI(api_key=os.getenv('OPENAI_API_KEY'))
    set_default_openai_client(client)
    
    # Instantiate the MCP server connection to Observee
    # Replace {your_client_id} with your actual client ID from the Observee dashboard
    mcp_server = MCPServerStreamableHttp(
        name='Observee',
        params={
            'url': 'https://mcp.observee.ai/mcp?client_id={your_client_id}',
            'headers': {
                'Authorization': 'Bearer {observee_api_key}'  # Replace with your API key
            }
        }
    )

    # Create the Agent and register the MCP server
    agent = Agent(
        name='Observee Agent',
        instructions='Use all available tools and knowledge to help users with their requests.',
        mcp_servers=[mcp_server]
    )
    
    try:
        # Use MCP server as async context manager
        async with mcp_server:
            # Example query with streaming - the Agent will use appropriate Observee tools
            result = Runner.run_streamed(
                agent, 
                'check my latest emails'
            )
            
            # Stream the response as it comes
            async for event in result.stream_events():
                if event.type == "raw_response_event" and isinstance(event.data, ResponseTextDeltaEvent):
                    print(event.data.delta, end="", flush=True)
            
            print()  # Final newline
            
    except Exception as error:
        print(f'Connection failed: {error}')
        # Check your client ID, API key, and network connection

if __name__ == "__main__":
    asyncio.run(main())

​
Environment Setup

Create a .env file with your credentials: 
OPENAI_API_KEY=your_openai_api_key_here

​
Learn More