Google Search Console
MCP server.
Turn search performance into your next SEO decision.
Give your AI client access to the Search Console properties your Google account can read. Explore the queries and pages behind your clicks, without copying reports into a chat. The hosted Google Search Console MCP server needs no API keys, no local process and no cloud project: paste one endpoint, sign in with Google, and ask.
https://mcp.liftaven.com/gscYour data. Better questions.
Find search opportunities
Compare clicks, impressions, click-through rate and position. Ask which pages have visibility but attract fewer clicks.
Investigate changes
Compare supported reporting windows, then break performance down by query or page. Keep the property and dates visible in the answer.
Check sitemaps
Read sitemap information and explicitly request sitemap submission. This connection does not edit your website.
Connect in three steps
- Add a remote MCP server. Use an AI client with remote HTTP MCP and OAuth support. Client availability and plan requirements vary.
- Paste your endpoint. Use
https://mcp.liftaven.com/gsc. Start the connection from your AI client, rather than opening the endpoint as a normal web page. - Authorise with Google. Review the requested access, finish consent and return to your client. Ask it to list your accessible properties or sites first.
Your first conversation
After consent, check that your client lists the server’s tools. Start by listing sites or properties, choose the correct one, then ask a question with an explicit date range. Ask the agent to include the source, dates and any missing data in its answer.
If the connection does not work
A 401 response when opening the endpoint in a browser is expected: connect from your MCP client so it can start OAuth. If your property is missing, check that you authorised the Google account with access to it. If consent expires, reconnect through your client. If the client does not support remote MCP with OAuth, use a compatible client rather than pasting access tokens.
You do not need to create your own OAuth app or paste a provider token. The client receives a scoped MCP connection. You can revoke access when you no longer need it.
Set up the GSC MCP server in your client
Every client below supports remote HTTP MCP with OAuth. Add the server, start the connection, finish Google consent in the browser and return to the client. The server name is yours to choose; liftaven-gsc is used in the examples.
Claude (web and desktop)
Open Settings, choose Connectors, then Add custom connector. Paste https://mcp.liftaven.com/gsc as the remote MCP server URL and save. Click Connect on the new connector to start Google sign-in. Availability depends on your Claude plan and organisation settings.
Claude Code
Run the command below in your terminal, then type /mcp inside Claude Code to authenticate.
claude mcp add --transport http liftaven-gsc https://mcp.liftaven.com/gscCursor
Add the entry below to .cursor/mcp.json in your project, or to the global MCP settings. Cursor opens the browser for OAuth the first time the server is used.
{
"mcpServers": {
"liftaven-gsc": {
"url": "https://mcp.liftaven.com/gsc"
}
}
}VS Code (GitHub Copilot agent mode)
Create .vscode/mcp.json with the entry below, then start the server from the MCP view.
{
"servers": {
"liftaven-gsc": {
"type": "http",
"url": "https://mcp.liftaven.com/gsc"
}
}
}ChatGPT and other clients
Use the client’s custom connector or remote MCP option where the plan allows it, and paste the same URL. If a client only supports local stdio servers, a bridge such as mcp-remote can forward to the hosted endpoint; native remote support is preferable when available.
What the GSC MCP tools return
Each tool returns a bounded, labelled report rather than a raw API dump, so an agent can cite the property, window and definitions it used.
list_propertiesEvery Search Console property your Google account can read, with the exact siteUrl to use in other calls.
get_performanceClicks, impressions, CTR and average position for a 7-day, 28-day or 3-month window, with the preceding period for comparison.
get_breakdownTop rows by query, page, country, device or search appearance, including previous-period clicks.
find_striking_distanceQueries ranking just off page one, sorted by impressions, so a small improvement can cross into the clicks.
inspect_url / inspect_urlsStored index status, canonical and crawl information for up to ten URLs at a time.
list_sitemapsSubmitted sitemaps with URL counts, errors and last download time.
submit_sitemapThe only write: register or re-fetch a sitemap you confirm. Submission is a request, not an indexing guarantee.
Start with a useful question
Example prompts. Your agent’s answers depend on the data and permissions available.
List the Search Console properties I can access.
For my website, compare the last 28 days with the previous period. Which pages lost clicks?
Find queries with strong impressions and low CTR. Explain the evidence before suggesting title changes.
Which queries sit between position 11 and 20 with real impressions? Group them by page.
Hosted GSC MCP server vs running your own
Open-source Google Search Console MCP servers on GitHub work well for developers who are comfortable creating API credentials and running a local process. The hosted endpoint trades that control for a two-minute setup and nothing to maintain.
| Consideration | Self-hosted open-source server | Liftaven hosted endpoint |
|---|---|---|
| Setup | Create a cloud project or app registration, enable APIs, generate credentials, install and run locally | Paste one URL into your client and sign in with Google |
| Credentials | Stored on your machine; you rotate and secure them | OAuth tokens held server-side, scoped to read access; revoke from your provider account at any time |
| Works from | The computer running the process | Any client on any device, including web-based clients |
| Write access | Depends on the project; some expose write tools | Read-only, plus an explicit sitemap submission you confirm |
| Reporting limits | Raw API limits | Bounded reports with labelled windows and row caps; Pro extends windows and rows |
| Cost | Free software; your time to set up and maintain | Free plan; Pro for longer windows |
Both approaches read the same provider data. Choose self-hosting if you need custom tools or an air-gapped setup; choose the hosted endpoint if you want your team using GSC data in their AI client this afternoon.
Access and limits
Start free with 7-day, 28-day and 3-month reporting windows and up to 25 rows per breakdown. Longer windows and larger breakdowns are Pro features. Results are bounded reports, not a full Search Console export.
Your AI client receives the reports it requests. Review its data policies before connecting. Read Liftaven’s privacy and permissions.
Google Search Console MCP server: questions, answered
Do I need to host a Google Search Console MCP server myself?
No. Add the hosted endpoint to a client that supports remote MCP and OAuth, then authorise with Google. There is no Google Cloud project, service account or JSON key to create.
Which Google account should I authorise?
The one that already has access to the property in Search Console. Every property that account can read becomes available immediately; there is no activation step.
Can the agent change my website or my Search Console settings?
No. This endpoint reads search reports and supports an explicit sitemap submission you confirm. It cannot delete sitemaps, add or remove properties, request indexing or change settings.
Is there an official Google Search Console MCP server from Google?
Not as of October 2026. Google publishes managed MCP servers for several Cloud and developer products, but Search Console is not among them. The Search Console MCP servers available today are community open-source projects you run yourself, or hosted endpoints such as this one that call the official Search Console API on your behalf.
Which Google Search Console MCP server should I use?
Choose an open-source server from GitHub if you are comfortable creating a Google Cloud project, enabling the Search Console API, creating an OAuth client or service account and running a local process; you keep full control and can add your own tools. Choose a hosted server if you want to paste one URL into Claude, ChatGPT or Cursor and sign in with Google, or if your client runs in the browser where a local process is not available. Both read the same Search Console data, so compare setup effort, write access and reporting limits rather than accuracy.
Do I need a Google Cloud project or API key?
No. The hosted server uses its own verified Google OAuth client. You sign in with the Google account that can already see your Search Console properties and approve read access; nothing is pasted into your AI client except the endpoint URL.
Why is average position different from a ranking?
Average position is an impression-weighted mean across every query in scope. It moves when the query mix changes even if no ranking did. Filter to one query before reading it as a rank.
Why do query rows not add up to the site totals?
Google withholds rare queries for privacy and de-duplicates impressions differently per dimension. Treat totals as authoritative and rows as a ranked sample.
Is there a free plan?
Yes. The free plan covers the 7-day, 28-day and 3-month windows with 25 rows per breakdown, which is enough for a weekly review of a small site. Pro adds 6- and 12-month windows and up to 200 rows.
Does this work with Claude, ChatGPT, Cursor and VS Code?
It works with any client that supports remote HTTP MCP with OAuth. Claude, Claude Code, Cursor and VS Code are covered in the setup guide above; ChatGPT needs a plan that allows custom connectors.
Need all your SEO data together?
The unified Liftaven MCP uses https://mcp.liftaven.com/mcp. Sign into Liftaven and approve your client to reuse the connections in your dashboard. Standalone provider endpoints have their own consent.