Gemini
Gemini doesn't yet ship an Anthropic-style native MCP connector, so the most reliable approach is to register the FindIP REST API as a tool via Function Calling in the Gemini API or Google AI Studio.
Prerequisites
A Gemini API key (issued from Google AI Studio) plus a FindIP API key.
Setup steps
Issue a FindIP API key
Create a key in the dashboard and store it as an environment variable.
export FINDIP_API_KEY="psk_live_..."
Write the Function Declaration
Declare FindIP search as a tool in the Gemini SDK.
from google import genai
from google.genai import types
client = genai.Client()
findip_search = types.FunctionDeclaration(
name="findip_search",
description="Semantic search across patents from KR/US/JP/CN/EP.",
parameters={
"type": "object",
"properties": {
"query": {"type": "string"},
"top_k": {"type": "integer", "default": 10},
},
"required": ["query"],
},
)
tools = types.Tool(function_declarations=[findip_search])Tool-call handler
When the model emits a function call, hit the FindIP REST API for real.
import os, requests
def handle_findip_search(args):
res = requests.post(
"https://api.findip.ai/api/v1/search/semantic",
headers={"X-API-Key": os.environ["FINDIP_API_KEY"]},
json=args,
timeout=30,
)
return res.json()Run the conversation
Send the natural-language query to Gemini → the model issues a tool call → return the result to the model → get the final answer.
Sample prompt
Prompt
"Summarize Korean filing trends in solid-oxide fuel cells (SOFC) over the past 3 years and list the top 5 applicants."
Troubleshooting
Can I register FindIP in Gemini Extensions?
Extensions only accept entries curated by Google. For custom integrations, use Function Calling or Agent Builder.
Does this work the same on Vertex AI?
Yes — register the same declaration on Vertex AI Gemini's Tool feature. Only swap the SDK to the vertexai package; everything else stays the same.