OpenAI MCP Integration

Guide for integrating OpenAI models with PraisonAI agents using MCP

MCP(...) accepts three equivalent construction forms — see Three equivalent forms.

Add OpenAI Tool to AI Agent

Quick Start

1. Set API Key

   Set your OpenAI API key as an environment variable in your terminal:

   export OPENAI_API_KEY=your_openai_api_key_here

2. Create a file

   Create a new file openai_airbnb.py with the following code:

   from praisonaiagents import Agent, MCP
   import os
   
   # Get API key from environment variable
   openai_api_key = os.environ.get("OPENAI_API_KEY")
   
   search_agent = Agent(
       instructions="""You help book apartments on Airbnb.""",
       llm="gpt-4o-mini",
       tools=MCP(
           command="npx",
           args=["-y", "@openbnb/mcp-server-airbnb", "--ignore-robots-txt"],
           env={"OPENAI_API_KEY": openai_api_key}
       )
   )
   
   search_agent.start("I want to book an apartment in Paris for 2 nights. 03/28 - 03/30 for 2 adults")

3. Install Dependencies

   Make sure you have Node.js installed, as the MCP server requires it:

   pip install praisonaiagents

4. Run the Agent

   Execute your script:

   python openai_airbnb.py

Gradio UI

from praisonaiagents import Agent, MCP
import gradio as gr
import os

# Get API key from environment variable
openai_api_key = os.environ.get("OPENAI_API_KEY")

def search_airbnb(query):
    agent = Agent(
        instructions="You help book apartments on Airbnb.",
        llm="gpt-4o-mini",
        tools=MCP(
            command="npx",
            args=["-y", "@openbnb/mcp-server-airbnb", "--ignore-robots-txt"],
            env={"OPENAI_API_KEY": openai_api_key}
        )
    )
    result = agent.start(query)
    return f"## Airbnb Search Results\n\n{result}"

demo = gr.Interface(
    fn=search_airbnb,
    inputs=gr.Textbox(placeholder="I want to book an apartment in Paris for 2 nights..."),
    outputs=gr.Markdown(),
    title="Airbnb Booking Assistant",
    description="Enter your booking requirements below:"
)

if __name__ == "__main__":
    demo.launch()

Features