> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agno.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Serve an Airbnb Specialist over A2A

> Connect an Agent to the OpenBNB MCP server and expose it on the dedicated A2A topology port.

Connect an Agent to the OpenBNB MCP server and expose it on the dedicated A2A topology port. AgentOS manages the MCP subprocess for the server lifespan.

```python airbnb_agent.py theme={null}
"""
Serve an Airbnb Specialist over A2A
===================================

Connect an Agent to the OpenBNB MCP server and expose it on the dedicated A2A
topology port. AgentOS manages the MCP subprocess for the server lifespan.

Prerequisites: OPENAI_API_KEY, Node.js with npx, internet access, and the `agno[a2a,mcp]` extras
Run: .venvs/demo/bin/python cookbook/05_agent_os/15_a2a/multi_agent/airbnb_agent.py
Try: Fetch GET http://127.0.0.1:7783/a2a/agents/airbnb-agent/.well-known/agent-card.json
"""

from agno.agent import Agent
from agno.db.sqlite import SqliteDb
from agno.models.openai import OpenAIResponses
from agno.os import AgentOS
from agno.tools.mcp import MCPTools

# ---------------------------------------------------------------------------
# Create Airbnb Agent
# ---------------------------------------------------------------------------

db = SqliteDb(
    id="a2a-airbnb-db",
    db_file="tmp/a2a_airbnb.db",
)

airbnb_tools = MCPTools(
    command="npx -y @openbnb/mcp-server-airbnb --ignore-robots-txt",
    timeout_seconds=30,
)

airbnb_agent = Agent(
    id="airbnb-agent",
    name="Airbnb Agent",
    description="An A2A specialist for Airbnb accommodation searches.",
    model=OpenAIResponses(id="gpt-5.5"),
    db=db,
    tools=[airbnb_tools],
    instructions=[
        "Call airbnb_search exactly once to find stays that match the request.",
        "Use airbnb_listing_details only when more detail is needed.",
        "If a tool returns an error, do not retry it; report the error and search URL.",
        "Return a short shortlist with dates, prices, and listing links.",
    ],
    markdown=True,
)

# ---------------------------------------------------------------------------
# Create AgentOS
# ---------------------------------------------------------------------------

agent_os = AgentOS(
    id="a2a-airbnb-os",
    description="AgentOS serving the Airbnb specialist.",
    agents=[airbnb_agent],
    a2a_interface=True,
)
app = agent_os.get_app()

# ---------------------------------------------------------------------------
# Run Airbnb AgentOS
# ---------------------------------------------------------------------------

if __name__ == "__main__":
    agent_os.serve(app=app, port=7783)
```

## Run the Example

<Steps>
  <Snippet file="create-venv-step.mdx" />

  <Step title="Install dependencies">
    ```bash theme={null}
    uv pip install -U "agno[a2a,mcp,os]" openai
    ```
  </Step>

  <Step title="Prepare Node.js">
    The MCP server runs with `npx`. Install Node.js, then verify the commands:

    ```bash theme={null}
    node --version
    npx --version
    ```
  </Step>

  <Step title="Export your OpenAI API key">
    <CodeGroup>
      ```bash Mac/Linux theme={null}
      export OPENAI_API_KEY="your_openai_api_key_here"
      ```

      ```bash Windows theme={null}
      $Env:OPENAI_API_KEY="your_openai_api_key_here"
      ```
    </CodeGroup>
  </Step>

  <Step title="Run the example">
    Save the code above as `airbnb_agent.py`, then run:

    ```bash theme={null}
    python airbnb_agent.py
    ```
  </Step>
</Steps>

Full source: [cookbook/05\_agent\_os/15\_a2a/multi\_agent/airbnb\_agent.py](https://github.com/agno-agi/agno/blob/v3.0.4/cookbook/05_agent_os/15_a2a/multi_agent/airbnb_agent.py)