Connect GSC to Gemini

How can I connect Google Search Console data with Gemini?

Gemini reads your Search Console data through the same MCP endpoint every other assistant uses. Today that means Gemini CLI, Gemini Enterprise or the Gemini API: point them at mcp.gscwizard.com and Gemini works from real clicks, impressions, position and CTR.

Get your API key View MCP docs and tools

Endpoint: https://mcp.gscwizard.com/mcp

The short answer

Add the GSC Wizard MCP server to whichever Gemini surface you use:

  1. Gemini CLI: one gemini mcp add command, or an mcpServers block in settings.json.
  2. Gemini Enterprise: register https://mcp.gscwizard.com/mcp as a custom MCP server connection.
  3. The Gemini API: pass an MCP client session into tools with the Google Gen AI SDK, or call the REST surface at /v1 directly.

The consumer Gemini app does not offer custom MCP connectors yet, so use one of the routes above.

Video

Watch the setup end to end

A walkthrough of the same MCP server Gemini connects to: adding it, authenticating, and asking for real Search Console analysis in plain language.

Option 1 - recommended

Gemini CLI

Gemini CLI speaks streamable HTTP and forwards custom headers, so the GSC Wizard endpoint drops straight in.

  1. 1

    Create an API key in GSC Wizard

    Sign in at tool.gscwizard.com, connect your Search Console properties, then create a key under Account → API keys.

  2. 2

    Register the server with Gemini CLI

    Run gemini mcp add, or add an mcpServers entry to ~/.gemini/settings.json with the endpoint and your bearer header.

  3. 3

    Verify and ask

    Type /mcp inside the CLI to confirm the server is connected and list its tools, then ask about your site in plain language.

One-line install

gemini mcp add --transport http \
  --header "Authorization: Bearer gscw_live_..." \
  gsc-wizard https://mcp.gscwizard.com/mcp

Or edit ~/.gemini/settings.json

{
  "mcpServers": {
    "gsc-wizard": {
      "httpUrl": "https://mcp.gscwizard.com/mcp",
      "headers": {
        "Authorization": "Bearer gscw_live_..."
      },
      "timeout": 30000
    }
  }
}

Use ~/.gemini/settings.json for every project, or .gemini/settings.json inside a repository to scope it there. Keep secrets out of source control by referencing an environment variable, for example "Authorization": "Bearer $GSCW_KEY", which the CLI expands at runtime.

Gemini CLI also supports OAuth for remote MCP servers. If you would rather sign in than store a key, omit the header and use /mcp auth to run the browser flow.

Option 2 - organisations

Gemini Enterprise

Gemini Enterprise supports custom MCP server connections, so an admin can make Search Console data available to agents across the organisation.

  1. 1

    Create a read-only API key

    In GSC Wizard, create a key scoped to read only under Account → API keys. A read key physically cannot call a write tool, which is usually what you want for a shared deployment.

  2. 2

    Add the MCP server connection

    In the Gemini Enterprise admin console, add a custom MCP server with the URL https://mcp.gscwizard.com/mcp and an Authorization: Bearer header holding the key.

  3. 3

    Share it with the right agents

    Grant access to the agents or spaces that should see Search Console data. Access stays scoped to the GSC Wizard account that issued the key, and revoking the key cuts it off immediately.

The same pattern works for other Google surfaces that accept MCP servers, including Gemini Code Assist and Gemini in Android Studio: the URL and the bearer header never change.

Option 3 - developers

The Gemini API and your own agents

Building an agent instead of chatting? The Google Gen AI SDK can pass an MCP client session straight into the tools list, so Gemini calls GSC Wizard tools directly.

Python: google-genai with an MCP session

from google import genai
from mcp import ClientSession
from mcp.client.streamable_http import streamablehttp_client

URL = "https://mcp.gscwizard.com/mcp"
HEADERS = {"Authorization": "Bearer gscw_live_..."}
MODEL = "gemini-3.1-pro"  # or whichever Gemini model you use

async with streamablehttp_client(URL, headers=HEADERS) as (read, write, _):
    async with ClientSession(read, write) as session:
        await session.initialize()
        response = await genai.Client().aio.models.generate_content(
            model=MODEL,
            contents="Which pages on example.com lost the most clicks in the last 28 days?",
            config=genai.types.GenerateContentConfig(tools=[session]),
        )
        print(response.text)

If you would rather not speak MCP at all, the same keys, scopes and rate limits back a plain REST surface at https://mcp.gscwizard.com/v1, with a generated OpenAPI 3.1 document at /v1/openapi.json that you can hand to Gemini as a function-calling spec.

What about the Gemini app itself?

The consumer Gemini app connects to Google's own apps, but it has no option to add a custom MCP connector, so a Search Console MCP cannot be attached there yet. Until that changes, use Gemini CLI, Gemini Enterprise or the API. If you only want to hand a colleague the numbers, a shared report link or a CSV export from GSC Wizard works in any chat.

Plain language in, finished answers out

What you can ask once it is connected

Every analysis runs server-side against the data warehouse, so the assistant reads a finished result instead of doing arithmetic over raw rows.

"Give me a performance summary for example.com" "What are my top queries and pages this month?" "Which queries gained or lost rankings versus last month?" "Find pages that are decaying" "Where am I cannibalizing keywords?" "Show my CTR versus position curve" "Generate a full SEO report" "Inspect https://example.com/pricing in Google" "Which pages are close to page one?" "Submit these URLs to IndexNow"
Same data, other assistants

Connecting a different AI tool?

One Search Console connection, every assistant. Pick your client and follow the step-by-step guide.

Connect GSC to ChatGPT

The GSC Wizard app, a custom connector, or the Responses API.

Read the guide

Connect GSC to Claude

A custom connector in the app, or one command in Claude Code.

Read the guide

Connect GSC to Grok

A custom MCP connector at grok.com/connectors, or the xAI API.

Read the guide

Connect GSC to any MCP client

The endpoint, both auth modes, and a config for every client.

Read the guide

See also the Google Search Console MCP overview, the GA4 MCP, and the ChatGPT app.

Frequently asked questions

Can the Gemini app connect to Google Search Console?

The consumer Gemini app has no add-a-custom-connector option, so a Search Console MCP cannot be attached there yet. Custom MCP servers are supported in Gemini CLI, in Gemini Enterprise, and through the Gemini API, and all three can reach GSC Wizard at https://mcp.gscwizard.com/mcp.

Does Gemini CLI support remote MCP servers with authentication?

Yes. Use the httpUrl key for a streamable HTTP endpoint and a headers object for the Authorization header, or run gemini mcp add with --transport http and --header. Gemini CLI also supports OAuth for remote MCP servers, managed with the /mcp auth command.

Where do I put the settings.json file?

Use ~/.gemini/settings.json for every project on the machine, or .gemini/settings.json inside a project to scope the server to that repository. Prefer an environment variable over a hardcoded key: Gemini CLI expands values such as $GSCW_KEY at runtime.

Is the data sampled or capped like a Search Console export?

No. Reads are served from a ClickHouse warehouse holding the full row set, so there are no export row caps and no GSC API paging. The warehouse lags roughly two days behind, and each response reports the date it is settled through.

What can Gemini do once it is connected?

Anything the MCP exposes: search analytics, top queries and pages, ranking changes, decay and cannibalization, CTR curves, opportunity scoring, bulk URL inspection, indexing tracking, IndexNow submissions, Bing Webmaster data, GA4 metrics, and full SEO reports.

Does it cost anything?

API keys are free for every GSC Wizard account, and the free tier works through the GSC API. Subscribers get the fast warehouse-backed querying, which is what makes full-dataset analysis practical. See the pricing page for details.

Point Gemini at your Search Console data

Create a free API key, add one entry to your Gemini config, and ask about your site in plain language.

Get your API key Read the MCP docs