Vertex AI

Native MCP in the Gemini SDK.

Build a Gemini app on Vertex? The Gemini Python/JS SDK has a native MCP client — pass the BankBridge URL in your tools list.

Step 01

Register BankBridge as an MCP tool.

Python

from google import genai
from google.genai import types

client = genai.Client(vertexai=True, project="your-project", location="us-central1")

response = client.models.generate_content(
    model="gemini-2.0-flash",
    contents="How much did I spend on restaurants in March?",
    config=types.GenerateContentConfig(
        tools=[
            types.Tool(
                mcp=types.McpTool(
                    endpoint="https://bankbridge.money/api/mcp",
                    headers={"Authorization": "Bearer ldgr_your_api_key_here"},
                )
            )
        ]
    ),
)
print(response.text)

Replace ldgr_your_api_key_here with a real key from your keys page. The same shape works in the Agent Development Kit (ADK) — any ADK agent can attach MCP tools this way.

Step 02

There is no step 2.

Vertex proxies the tool calls. Your app sees the model’s final answer.

Problems? Email hello@bankbridge.money.