Prompt Research, AI Deep Dives and MCP Access for all: SISTRIX Updates in the Second Quarter of 2026

From:

SISTRIX Content Team

Published: 31.07.2026

AI Overviews and Chatbots always answer a question directly, which means there are more and more zero-click-searches. At position 1, the click-rate of a search with an AI Overview is reduced from 27% to 11%. For SEO teams the classical ranking is now less important than the question if and how a brand is mentioned (Mentions) or cited (Citations) in AI responses. Visibility has to be re-thought and this requires reliable data. The updates introduced in Q2 of 2026 tackle this exact problem.

SISTRIX for AI/Chatbots

A distinctive focus during the second quarter was on the visibility in AI systems, especially regarding Prompt Research and the new AI Deep Dives.

Prompt Research: the Keyword Research for AI Search

In the conventional Google search, demand was clearly measurable based on keywords and search volume. For AI search, this is not as easy anymore: users formulate individual, often one-of-a-kind prompts, that can barely be analysed individually.

Prompt Research solves this by bundling similar prompts into a topic. The basis is made up of over 62 million real user questions for more than 1.4 million topics. Per topic, aggregate values are created such as the new Search Volume for the AI search, Intent, and Customer Journey Stage. The detailed view offers a complete content briefing with buyer persona, emotional driver, unposed questions and platform-specific user prompts.

This way, for the first time AI search requests are systematically analysable and directly usable as a briefing. You can find further details in the handbook.

Prompt Tracking: AI Tag Suggestions and Citations

The individual Prompt Tracking was extended as well. The thematic structuring of prompts had to be done manually so far. With the new function “Tag Suggestions”, SISTRIX analyses the prompts now with AI, recognises topics and suggests fitting tags, including automatic classification. In the new “By Tag” view, prompts are grouped and displayed alongside metrics such as Visibility Score and Share of Voice, compared with the previous period.

Another new feature is the ability to link domains to brands. This means that, for the first time, not only Mentions but also Citations can be measured. Precisely because the sources cited in AI responses change significantly from week to week, it is worth monitoring your own citation presence over time. We have summarised a detailed explanation of the various new features in the changelog.

AI Deep Dives: SISTRIX Data meets AI Interpretation

A generic AI model can read and interpret data, but it cannot reliably calculate. With a thin base it invents facts and reputable expert knowledge can barely be distinguished from mere claims.

So far, the combination of AI interpretation and AI data could mostly be used in AI systems. With the AI Deep Dives, you can now do the same thing directly in the Toolbox: the AI reads and interprets and the deterministically calculated numbers, as well as the tested SEO and GEO knowledge comes from SISTRIX. The analyses can be created with the press of a button in the filter bar of the corresponding function.

To start with, there are seven Deep Dives in the AI Check and in the Prompt Research which you can download as Markdown or PDF respectively.

This way, there are no made-up numbers but proven findings that are comprehensible without deeper SEO or GEO knowledge. Whether the analysis is created for the brand sentiment, buyer types or concrete content chances: every Deep Dive delivers directly usable results instead of a pure data table. Additional details, explanations and the individual AI Deep Dives are available in the changelog.

MCP Server: OAuth now for everyone

The SISTRIX MCP Server is now, with the switch to OAuth, available to all SISTRIX users without the need for an API key. If you have an account, log in and get started immediately.

Via the Model Context Protocol, chatbots such as ChatGPT and Claude can get direct access to current SISTRIX data. The chatbot thus turns into a direct interface to SISTRIX: questions like “How has the Visibility Index of example.com developed?” are answered without a manual request or tab change. We have also further examined how exactly this works and initial use cases.

The deciding advantage: by using the OAuth access, the usage of the MCP server does not use any API credits. Whoever prefers to continue working with the API, can continue to do so per API Key.

SISTRIX API

Via the SISTRIX API, SEO and GEO numbers can be automatically retrieved and included into your own dashboards and reporting tools. In the second quarter, multiple new developments have been added that make it easier to integrate SISTRIX data into your workflows.

New Functions: domain.paths and domain.hosts

Until now, visibility data for the directory or host level could only be viewed directly in SISTRIX. But via domain.paths and domain.hosts, the best-ranking directories and hosts of a domain can be retrieved per API with their corresponding number of Top-10 and Top-100 keywords, as well as their share of the Visibility Index of the domain. This is equivalent to the directory and host tables in SISTRIX.

In effect, this means you can see at a glance how visible singular areas of a website are, like the directory /blog/ or a subdomain such as shop.example.com, and automatically add these values into your own reporting.

The Comeback of domain.urlcount.seo

The function is back. Via the end point domain.urlcount.seo, the Top-10 and Top-100 URL count of a domain, host, or path can be retrieved. To simplify: you can see how many URLs of an area of a website rank in the Google Top 10 or Top 100. A good indicator to discover how widely a directory or subdomain is represented in the search results.

Retrospective: the Biggest Innovations from the First Quarter of 2026

In the first quarter of 2026, a lot revolved around a changing search reality as well, from AI visibility to the Onpage Analysis and the API. As a reminder, here are once again the most important updates from Q1 summarised:

1. A reworked main navigation bundles all areas into a central dropdown and is available in the SISTRIX Labs to test out.

2. 

3. The updated header is a SISTRIX Labs Feature. This means you will have to activate it there first to try it out.
4. In the Prompt Tracking, you can now monitor the Google AI Mode next to ChatGPT, Perplexity and the Google AI Overviews.
5. An improved brand recognition ensures more precise data and less wrong allocations in AI responses.
6. The Onpage Projects were largely reworked, among other things with tag-based comparisons in the Project Visibility Index, new error identification and a livelog during the crawl.
7. The new Looker Studio Connector integrates the entire SISTRIX API, including AI visibility data, per no-code into Google Looker Studio.

Test All Updates from Q2 2026

The updates from the second quarter of 2026 adjust SISTRIX to fit the changing reality of search. AI Overviews and chatbots are now a fixed element of the work of those responsible for SEO. While Google continues to increasingly deliver responses themselves, SISTRIX makes it possible to measure who is named as a source and cited.

All new functions are now available to all SISTRIX users.

As always, we are happy to receive feedback! Praise showcases that we’re on the right track and critical feedback helps to improve SISTRIX even further. If you have questions or suggestions to new or existing functions: our support team will be glad to help at support@sistrix.com.