google_search
Performs a web search using Google Search (via the Gemini API) and returns the results. This tool is useful for finding information on the internet based on a query.
LABSCO SUMMARY
No second search provider to sign up for, and no second key to hold.
The whole server is one tool with one parameter, which makes it the cheapest way to get a model off its training data and onto today's web. It returns search results, not page bodies — pairing it with something that fetches a URL is the obvious next step.
What it is
A single-tool server: google_search takes a query string, runs a web search through Google Search via the Gemini API, and returns the results.
What you get
- Web search from one query parameter, answered by Google Search through the Gemini API
- One credential for both the model and the search — a Gemini API key in GEMINI_API_KEY, or on Vertex AI your own Google Cloud project (GEMINI_PROVIDER=vertex plus VERTEX_PROJECT_ID) instead of a key
- One entry in the client's tool list, which is the entire integration surface to review
Requirements
A Gemini API key in GEMINI_API_KEY — or, on Vertex AI, a Google Cloud project with Vertex AI enabled, named in VERTEX_PROJECT_ID with GEMINI_PROVIDER=vertex — and npx on your PATH.
Setup effort
One command plus a key — npm install -g mcp-gemini-google-search, then supply credentials
USE CASES
01
Answering a question that postdates the model
A release, an outage, a version number changed after training. One query goes out to Google Search through the Gemini API and the results come back into the conversation.
02
Checking a claim before it goes in writing
Search the specific wording rather than trusting recall, and read what comes back before the sentence ships.
1 tool
Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.
BEFORE YOU INSTALL
What to know first
Tools you get
1 tool when we connected. GEMINI_MODEL picks the model (default gemini-2.5-flash) and GEMINI_PROVIDER switches between Google AI Studio and Vertex AI. Results come back as prose with numbered source citations.
Requires
- npx on PATH
- Node.js 18 or later
Where it runs
All 4 documented setups run the server on your machine.
CONTEXT COST
Light Standing cost
Rated by the size of the one tool definition that rides along on every turn
INSTALLATION
Connect your client
Server setup
TERMINAL
npm install -g mcp-gemini-google-searchMCP Clients
Claude Code — AI Studio, user
CLAUDE CODE — AI STUDIO, USER · TERMINAL
claude mcp add gemini-google-search -s user -e GEMINI_API_KEY="<YOUR_GEMINI_API_KEY>" -e GEMINI_MODEL="gemini-2.5-flash" -- npx mcp-gemini-google-searchClaude Code — AI Studio, project
CLAUDE CODE — AI STUDIO, PROJECT · TERMINAL
claude mcp add gemini-google-search -s project -e GEMINI_API_KEY="<YOUR_GEMINI_API_KEY>" -e GEMINI_MODEL="gemini-2.5-flash" -- npx mcp-gemini-google-searchClaude Code — Vertex AI, user
CLAUDE CODE — VERTEX AI, USER · TERMINAL
claude mcp add gemini-google-search -s user -e GEMINI_PROVIDER="vertex" -e VERTEX_PROJECT_ID="<YOUR_VERTEX_PROJECT_ID>" -e VERTEX_LOCATION="us-central1" -e GEMINI_MODEL="gemini-2.5-flash" -- npx mcp-gemini-google-searchClaude Code — Windows
CLAUDE CODE — WINDOWS · TERMINAL
claude mcp add gemini-google-search -e GEMINI_API_KEY="<YOUR_GEMINI_API_KEY>" -- cmd /c npx mcp-gemini-google-search
