The Fullstory MCP connector for Claude, with Google ranks and AI answer visibility added

The connector's listing shows 29 tools: 15 reads, 9 writes and 5 it does not classify. The reads include get_view_counts, get_funnel, get_metric, get_segment, get_sessions and get_session_events. The writes include build_segment, build_metric, build_funnel and the update tools, which save definitions inside Fullstory.

Fullstory's listing says it autocaptures clicks, page views, errors and frustration signals such as rage clicks and dead clicks, so many questions need no instrumentation. Every metric can be unpacked into the sessions behind it. That is a strong view of what happened on a page, and it starts after the visitor has already arrived.

Fullstory's developer guide has two prerequisites: Model Context Protocol enabled for your org, and StoryAI features enabled. If your org has opted out of StoryAI, the guide says your client will appear to connect, with OAuth completing and the server showing as connected, but no tools will be available. An account admin can check this under Settings, StoryAI Features.

Fullstory is a verified connector in Claude's connector directory. In Claude, open Customize, then Connectors, then Add, then Browse Connectors, search for Fullstory and click Connect. The guide says the directory connector installs the tools only, not Fullstory's separate skills. For tools without a browser sign-in, such as a shared gateway, it describes API key authentication, and its recommended key level is Standard. The mcp.so listing notes that MCP access is available on paid plans and not included in free plans.

Fullstory is the behavior side. Linkeddit is the search-visibility side, on one plan of $60 a month. Its search_visibility_get tool returns, for each tracked keyword, your best Google position, the change since the last check, whether an AI Overview showed and whether it listed your domain. Its answer_results_get tool reports presence, recommendation, share of voice, position, sentiment and citation share across the engines.

Linkeddit never connects to your Fullstory account and Fullstory never sees your keywords. Claude carries only the numbers it needs from one tool to the other, inside your own session. Linkeddit writes no advice, so Claude does the joining and the writing. For this workflow, tell Claude to use Fullstory's read tools only, so it does not save metrics or segments you did not ask for.

The visibility gap is a real complaint. A SaaS marketer on Reddit wrote: "In all possible keywords where our page is ranking number 1 to 3 on Google, we are not showing up on Google AIO, as if we got shadow banned." Rank in a behavior tool never shows that. A rank check next to the overview flag does.

On Linkeddit's test project, plausible.io, search_visibility_get returned 25 tracked keywords, 12 of them checked on 2026-10-28. Five of those 12 ranked in the top 10: google analytics alternative at position 2, matomo alternative at 2, privacy friendly analytics at 3, simple analytics alternative at 3 and plausible vs fathom at 5. The other seven were not in the top 10.

Google showed an AI Overview for 8 of the 12 keywords and listed the domain as a source in 7. Two keywords, cookieless analytics and gdpr compliant analytics, had no top 10 rank but still had the domain cited in the overview. The run was read only and the Fullstory side was not run.

Every row was a first check, so no change since the last check could be shown. To finish the report, ask Claude to take the landing pages behind those keywords, count their views with get_view_counts, and read a handful of sessions with get_sessions and get_session_events. A page that ranks and gets cited but loses visitors to a form error is a different problem from one that ranks nowhere.

Treat behavior and answers as two measurements of one change. After you publish a fix, Claude can ask Linkeddit's answer_verify_start to re-check the buying questions the change targeted across the five engines, then read the result with answer_verify_get. Starting a check spends credits, so Linkeddit returns an estimate first and Claude waits for your yes.

Then ask Fullstory for the same page before and after the publish date. Fullstory's release impact workflow compares a signal in the window before and the window after a ship, using discover_groups with compare_to_previous, and it applies to rage clicks, errors, drop-off or conversion. If the engines now name you and the page's drop-off fell, the change worked on both counts. On the test project, the three most recent audit runs all finished partial, which a good report states.

Fullstory's listing says you sign in with OAuth and that Claude acts with your permissions, seeing only what your account can see. Session data can include what visitors typed and clicked, so tell Claude which pages and date range it may read, and keep Fullstory's own privacy controls on.

Because the connector can also write, set Claude's tool approvals so that build and update calls ask first. Then read the answer against the numbers in Fullstory itself, and open the session link Fullstory returns for any claim you plan to act on.