Google has released two news, unnoticed : an MCP for Chrome DevTools and a MCP Google Analytics.

They add up to the stack of servers MCP already available.

Once, it is an accident ; twice is a coincidence ; three times, it is a trend…

If you are an SEO / GEO Consultant, and you don’t play already with the MCP (Model Context Protocol), you are losing big time.

MCP (Model Context Protocol), what is it ?

The MCP is a standard open-source (powered by Anthropic, business mother of Claude). A MCP acts as a ‘USB cable universal’ for theArtificial Intelligence.

It allows you to connect nomatter what language model (LLM) to local data (your files), or remote tools (Google Analytics, Chrome in this case) in a secure manner and standardized.

En what the MCP is revolutionary ? 

Prior to the MCP, the AI were isolated in a chat window. With the MCP, they become Agents capable ofacting on your infrastructure.

The 3 key points to remember :

• Standardization : need More to create a specific integration for each tool. An MCP server works for Claude, Cursor, Windsurfing, and tomorrow, Gemini.
• Connection to the Real (Ground Truth) : THEAI is not responding only with its ‘data fortraining’ (often outdated), but with your actual data and fresh (Live Data).
• Safety : Cis you (the client) that control what data theAI can see and what actions it can perform. TheAI does not ‘scrape’ not random, it uses a protocol that is allowed.

I’m not going to be alarmist, but you will need to install both MCP that I present to you below. Not for looks, but to understand how it works, and especially to improve your everyday life.

Chrome DevTools MCP

As an SEO, GEO Technical consultant , I must admit that I’m amazed by this MCP.

Before, the LLMs ‘hallucinated’ a lot the rendering of a page or content of a curl basic HTML.

With the MCP Chrome Dev Tools, you connect theAI directly to the debugger of Chrome via Puppeteer.

also, Dan Petrovic from Dejan.’ve conducted tests in which he explicitly asked ChatGPT to describe what it ‘ saw ‘ on a page. It has been demonstrated that ChatGPT (via OAI-SearchBot) acts as a collector of text (similar to a curl command) that as a browser.

How does the MCP Chrome Dev Tools ?

The MCP Chrome Dev Tools works as follows :

1. Your AI (in your code editor or terminal) sends a request to the natural language (ex: ‘Check why the site is slow’).
2. The MCP Server translates this in technical orders via the protocol Chrome DevTools Protocol (CDP).
3. Chrome (headless or visible) performs theaction (click, scan the network, takes a screenshot).
4. The result is returned to theAI that can then analyze the real data (logs, loading time).

PS : personally, I prefer to see the MCP to launch an instance of Chrome.It reminds me of the Google Computer Use Model, which I tested.

In my case, it was not crazy : impossible to pass the Captcha. I surely missed a trick, but the experience was enough for me to see the potential… and the current limitations.

Advantages of the MCP Chrome Dev Tools

• Console & Debugging : THEAI has access to the logs of the console (Console.logs, errors, JS). She can see theexact error that occurs at runtime.
• Inspection DOM : THEtool allows theAI to make ‘snapshots’ of the DOM to understand the actual structure of the page after the render JavaScript.
• Network (Network) : THEAI can monitor HTTP requests, identify the errors, 404 or 500, and analyze the problems of the HORNS, without you having to copy-and-paste the logs.
• Scripting : THEAI can execute JavaScript directly in the page (evaluate_script) to test patches live.
• Real Performance (Core Web Vital pieces) : THERN may initiate traces of performance to measure the LCP (Largest Contentful Paint) and the CLS (Cumulative Layout Shift) in real-time.
• Check the output : Ideal to check how Googlebot sees your site (SSR vs CSR). TheAI can tell you if the content is loaded in the DOM.
• Analysis of the tags are : You can ask theAI to ‘scan all the tags meta and canonical to the current page’ to verify compliance SEO technique instantly.

Howto install the MCP Chrome Dev Tools ?

• Pre-requisite : Node.js (v20+).

• In your config file (MCP for Claude Desktop or Cursor), add this :

JSON

{

  "mcpServers": {

    "chrome-devtools": {

      "command": "npx",

      "args": ["-y", "chrome-devtools-mcp@latest"]

    }

  }

}

Personally, I have it installed it on VS code. I prompt it directly and get responses in the form of page Markdown. I find it super convenient.

Case ofuse of the MCP Chrome Dev Tools for SEO / GEO

Once connected, here’s what you can ask your Agent (and whatit actually uses) :

1. Audit Rendering JS (Maps_page + evaluate_script)
   • The Prompt : ‘Go to this URL, wait until the network is quiet, and fetch me the H1 and Meta Description such thatthey are rendered in the DOM final.’
   • LInterest : Check if your Framework JS (React/Angular) fair SEO client-side. Cis the ultimate test of ‘Renderability’.
2. Audit Core Web Vital pieces (performance_start_trace)
   • The Prompt : ‘that Starts a trace of the performance on the Homepage. Parses the JSON result and tell me what script is blocking the Main Thread, and impact theINP.’
   • Power : THEAI does not give you just the score Lighthouse. She reads the stack trace and point the file .jscoupable.
3. Debugging Errors Gifs (get_console_message)
   • The Prompt : ‘Walk the tunnel ofpurchase. Is it thatthere are 404 errors or mistakes JS in the console when I click on Add to cart ?’
   • The GEO : If theAgent is not able to click becausea script crashes, the Search Bot AI will not be able to index the content behind.

Barriers to the installation of the MCP Chrome Dev Tools

I do not hide from you that I’ve galèré to install theMCP.

Cis not (yet) the ‘Plug & Play’. To have wiped plasters, here are 3 obstacles that you will encounter :

• The Context Token : Chrome DevTools extract the DOM is complete. If you ask Claude danalyse a page is too heavy, you want to explode your window context (and your credits API if you’re not in the unlimited plan). My advice : Ask theagentto extract selectors specific (ex: document.querySelector(hand)) rather than the entire page body.

• The ‘Headless’ Blockers : Even with the MCP, some of the sites (Cloudflare, Akamai) may detect that cis a Chrome ‘driven’ and block you. Thisis not a magic wand for scraper Amazon.

• ThePython environment/Node : If you have a conflict of versions on your machine (which happens to 99% of devs), theinstallation via uvx or npx may fail. Make sure youhave a clean environment.

MCP Google Analytics MCP

Not gonna lie, the interface of GA4 is a hell UX. We are allagreed on it. The MCP server allows you to bypass theinterface to talk directly to theAPI Data.

I’ve heard of Googlers to Search Central Zurich say they hate Google Analytics, and that they are sick of it in front of everyone. You can quote me on that 😂 I take responsibility , with supporting evidence…

How does the MCP Google Analytics ?

The server is written in Python. It orchestre two libraries official Google that exist for a time, but which were reserved for the developers backend :

1. google-analytics-data (Data API v1beta) : Cis the engine. Cis it that will search for the numbers.
2. google-analytics-admin (Admin API v1beta) : Cis the configurator. It is used to list the accounts, properties, and check the rights.

The Data Stream (The Pipeline)

Here’s what happens when you ask Claude : ‘How visitors yesterday ?’

1. Translation (LLM -> MCP) : the prompt is converted by the client MCP (Claude/Cursor) in a function call, standardised, type :JSON{ 'name': 'run_report', etc.}
2. Execution (Python Wrapper) : the MCP server receives this JSON. It instantiates the client BetaAnalyticsDataClient of the library google-analytics-data.
3. Authentication  : the script does not ask you for your password. He is looking for the ADC (Application Default Credentials).
4. API call & Response : Google returns a JSON gross. The MCP server cleans (to get rid of the unnecessary noise), and returns it to theLLM in the form of text or table-structured.
5. Interpretation : Claude receives the raw data (ex: {'activeUsers': 450}) and form the sentence for her : ‘did You have 450 visitors active yesterday.’

Chow to install the MCP Google Analytics ?

• Pre-requisites : Python 3.10+ and uv or pip.
• LAuthentication : cis where many go to block. It is necessary to configure the ‘Application Default Credentials’ (ADC) on the Google cloud. (I’m not going to lie to you, cwas painful).
• Config MCP

SON

{

  "mcpServers": {

    "analytics-mcp": {

      "command": "uvx",

      "args": ["analytics-mcp"],

      "env": {

        "GOOGLE_CLOUD_PROJECT": "votre-project-id"

      }

    }

  }

}

Ditto for Google Analytics MCP. I used it on VS Code.

Obviously, thisis not for everyday wear. But for someonewho is managing a lot of accounts on GA4, cis a life saver…

Use of the MCP Google Analytics for SEO

1. run_report (classic)
   • TheAI builds the request JSON to theAPI. You can ask : ‘Give me 10 pages with the highest rate ofengagement from the organic traffic in the last month.’
   • TheAngle GEO : Identify the content which ‘hang’ really theuser, because those are the models that AI is possible in their responses (User Signal).
2. run_realtime_report
   • Ideal for monitoring a product launch or a live migration without page refresh Realtime of GA4.

Combine the MCP GA4 and the MCP Chrome DevTools

Theoretically, one can combine the two MCP Analytics, and Chrome DevTools to hang out with insights nice.

Example : Analyze why a specific page doesn’t convert.

J'ai besoin d'une analyse croisée pour la page "https://aioseo.fr/en/top-10-tools-geo-to-track-your-position-ia-2025/" :
1. Utilise le MCP Google Analytics pour me donner le "Taux d'engagement" et le "Temps moyen" sur cette page les 28 derniers jours.
2. Utilise le MCP Chrome DevTools pour visiter cette même page. Fais un screenshot du premier écran (above the fold).
3. Analyse le screenshot : est-ce que le CTA principal est visible sans scroller ? Est-ce qu'une pop-up cache le contenu ?
4. Dis-moi si le problème visuel détecté peut expliquer la métrique GA4.

Crash Test’ GEO : is Your website ready for the LLMs ?

The MCP with ChromeDevTools, (Analytics a little less), allow us to make a lot of test GEO / AEO / LLM.

Test theaccessibility LLMs

Ask your Agent for MCP : ‘Go on the site, is the cheapest product in the category Shoes, and tell me sit’s in stock.’

• If theAgent is successful : Your internal mesh and your semantic tags are clear.
• If theAgent fails : It will often tell you ‘I can’t find the button for ‘filter by price’ or ‘The menu is empty’. Cis exactly the experience SearchGPT.

Test ofinteraction JS 

Ask : ‘Click on the button Add to cart. What is the response from the server (Network tab) ?’ A lot of sites React/Vue.js fail because theevent onClick is obscured. If theAgent is not able to click it, it will never be able to finalize an action for theuser (the future of search, transactional).

The rule ofthumb : If Claude via MCP can not do it, autonomous agents will not send you qualified traffic.

Recap on the MCP Chrome Dev Tools and MCP Google Analytics for GEO and LLMs

Name of the MCP Server	Technology / Language	Key Features	Case dUse SEO/GEO	Pre-requisites ofInstallation	Key Benefits
Chrome DevTools MCP	Node.js (v20+), Puppeteer, Chrome DevTools Protocol (CDP)	Inspection DOM, log, console, network monitoring (HTTP), execution of JS scripts (evaluate_script), captures dscreen, a measure of Core Web Vital pieces (LCP, CLS).	Audit rendering JS (SSR vs CSR), check the meta tags/canonical, debugging oferror 404/500, testaccessibility for the agents AI and bots research.	Node.js (v20+), command npx -y chrome-devtools-mcp@latest in the config file (MCP).	Prevents the hallucinations of LLM on the rendering, allows theAI dto analyze the real data (Ground Truth) and the errors ofa live performance.
Google Analytics MCP	Python (3.10+), google-analytics-data (v1beta), google-analytics-admin (v1beta)	Extraction of reports via the API Data (run_report), reports real-time (run_realtime_report), management of accounts and properties via API Admin.	Identification of the content to its strong commitment to optimizing GEO, monitoring of migration, cross-sectional analysis, visual performance / metrics GA4.	Python 3.10+, uv or pip, or configuration of the Application Default Credentials (ADC), and project ID for Google Cloud.	Bypasses theinterface GA4 complex, allows you toquery the data in natural language, and to clean the JSON raw for the LLM.

A word about Security of Data

Everything you do to pass an MCP is not confidential.

The results of your queries, the JSON of GA4, the HTML of a page inspected via DevTools, are sent to Anthropic or OpenAI for interpretation.

Therefore, never MCP on the accounts or projects that contain sensitive data or personal (health, identity specific) without anonymize before.

Because yes, the data will pass through the servers of the model.

Curiosity, yes.

Unconsciousness, non.

My Opinion on the MCP Chrome Dev tools and MCP Google Analytics

Google does not gifts. Out of these MCP, they say : ‘Here are the standards to speak with our tools’.

TheAPI GA4 to the truth of the figures.

Chrome DevTools for the truth of the rendering.

Install. Play with. Cis the best way to understand how the AI can ‘see’ is actually the web. And remember : Good GEO is good infrastructure.

• How to optimize for Google AI Overviews
• Query Fan out, the tech behind’VE Overviews and IA Mode
• Reciprocal Rank Fusion (RRF): How LLMs Rank Answers (ChatGPT, Gemini)
• SEO Quiz : do you know these SEO experts ?
• Google Search Central 2025 Zurich