Google Search Console

This connector wraps the Google Search Console API (searchconsole.googleapis.com): Search Analytics (clicks, impressions, CTR, average position), the URL Inspection API (index status, canonical, crawl, rich results), sitemaps and property management. 11 tools.

Start with gsc_seo_playbook. It costs no API call and returns ready-made recipes (period comparison, striking-distance keywords, cannibalisation, traffic-drop diagnosis, indexing audit…) with the exact request bodies to send.

Setup (one-time, about 10 minutes)

1. Open https://console.cloud.google.com, pick or create a project, then APIs & Services → Library → enable Google Search Console API.
2. APIs & Services → OAuth consent screen: choose Internal if you are on Google Workspace (simplest), otherwise External. For External, set the publishing status to In production: while an app is in Testing, Google revokes its refresh tokens after 7 days and the connector stops working every week. For your own use, the "unverified app" warning can be clicked through.
3. APIs & Services → Credentials → Create credentials → OAuth client ID → application type Web application → add the authorized redirect URI <your AnythingMCP URL>/api/mcp-oauth/callback (on AnythingMCP Cloud: https://cloud.anythingmcp.com/api/mcp-oauth/callback).
4. Paste the client ID and secret into GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET and install the connector.
5. On the connector page, click Authorize with Provider and sign in with a Google account that has access to your Search Console properties. The refresh token is stored encrypted and access tokens are renewed automatically.

Scope: the connector asks for https://www.googleapis.com/auth/webmasters (read and write), because it can also submit sitemaps and add properties. For a read-only setup, change the scope to https://www.googleapis.com/auth/webmasters.readonly in the connector's OAuth settings, re-authorize, and disable gsc_submit_sitemap, gsc_delete_sitemap, gsc_add_site and gsc_remove_site.

Properties (site_url)

• Always call gsc_list_sites first and reuse the siteUrl exactly as returned. https://example.com/, https://www.example.com/ and sc-domain:example.com are three different properties.
• A domain property (sc-domain:) covers every protocol and subdomain. To look at one subdomain or folder, keep the domain property and filter on page.
• permissionLevel siteUnverifiedUser means no data access: every data call on that property returns 403.

Search Analytics (gsc_query_search_analytics)

• startDate and endDate are required, YYYY-MM-DD, in Pacific Time. There are no relative dates: work them out from today's date. History goes back 16 months.
• Fresh data: dataState: "all" includes the last 2-3 days, which are still incomplete (what the Search Console UI shows). "final" returns only finalised days. "hourly_all" together with the hour dimension gives hourly data for about the last 10 days.
• Dimensions: query, page, country, device, searchAppearance, date, hour. With no dimensions you get one totals row.
• Rows come back sorted by clicks, descending (by date when date is a dimension). The API cannot sort by impressions, CTR or position: raise rowLimit (max 25,000 per call, page with startRow) and sort the result yourself.
• ctr is a fraction (0.034 = 3.4%). position is the average rank, 1 = top.
• Filters: dimensionFilterGroups: [{"groupType": "and", "filters": [{"dimension": "page", "operator": "contains", "expression": "/blog/"}]}]. Operators: equals, notEquals, contains, notContains, includingRegex, excludingRegex (RE2 syntax). Countries are ISO 3166-1 alpha-3 in lower case (ita, deu, usa); devices are DESKTOP, MOBILE, TABLET.
• Anonymised queries are left out of rows grouped by query, so their sum is lower than the property total. For site totals, query without dimensions.
• type: web (default), image, video, news, discover, googleNews. Discover and Google News have no query dimension and no position.

URL Inspection (gsc_inspect_url): one URL per call, and the URL must belong to the property. Quota: 2,000 inspections per day and 600 per minute per property. It returns Google's indexed view (verdict, coverage state, last crawl, Google-selected vs declared canonical, robots.txt, fetch state, rich results) plus inspectionResultLink to the report in the Search Console UI. It does not trigger a recrawl: the API has no "request indexing" call.

Sitemaps: gsc_list_sitemaps shows each sitemap's last submission and download, pending state, errors, warnings and submitted URL counts. The indexed counts in contents are no longer populated by Google; use URL Inspection or the page-indexing report for index coverage.

Errors: 403 means the Google account has no access to that property (or a read-only scope for a write call); 404 on a property usually means site_url is not spelled exactly like in gsc_list_sites; 429 means the per-minute quota, so wait and retry.