Umami Cloud MCP. Pull real-time, privacy-first web metrics into your prompts.
Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Umami Cloud is a privacy-focused web analytics MCP Server that lets your AI client pull real-time and historical site metrics directly into chat or code.
It tracks pageviews, active users, unique visitors, and device breakdowns without sacrificing user anonymity.
What your AI agents can do
Users
Gets the current count of active users on your website.
Websites
Retrieves specific metrics, including URLs viewed, browsers used, operating systems, and device types.
Instantly retrieve a live count of how many users are currently active on your site.
Get detailed breakdowns of traffic by URL, operating system, browser type, and device category (mobile/desktop).
Pull lists of top-performing pages, allowing you to see which URLs generate the most views over a specified time frame.
Query historical data to identify and rank major referral traffic sources (e.g., social media platforms, search engines).
Calculate key metrics like bounce rate or average session duration for specific segments of users.
Ask AI about this MCP
Supported MCP Clients
Waiting for input…
Umami Cloud MCP Server: 2 Tools for Web Analytics
These two tools allow your AI client to access core web analytics data, including active user counts and detailed site performance metrics, all through the Umami API.
019d7618users
Gets the current count of active users on your website.
019ddc1cwebsites
Retrieves specific metrics, including URLs viewed, browsers used, operating systems, and device types.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Umami Cloud, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
You're tired of hopping between dashboards, right? You gotta pull data from three different places just to get a basic idea of how your site's doing. This isn't that mess. The Umami Cloud MCP Server connects any agent you use—Claude, Cursor, whatever—straight to the core metrics without making you touch a dashboard or worry about tracking cookies.
It lets your AI client pull real-time and historical website data directly into your chat window or straight into your code base.
When you need to know how many people are actually on your site right now, you just call the users tool. It spits out an immediate, live count of active users. That's it—a single number telling you exactly how much traffic you've got running right this second. You don't have to guess; you get the hard metric instantly.
If you need a deep dive into what your site is actually doing, you use the websites tool. This thing pulls in a massive amount of detail about user behavior that goes way beyond simple visitor counts. It gives you detailed metrics on every URL viewed and tracks which browsers people are using.
You can see exact breakdowns of operating systems—whether it's iOS or Windows—and what kind of devices they’re hitting your site with, whether that's a mobile phone or a desktop machine.
The websites tool handles the heavy lifting for performance tracking. It lets you pull lists of top-performing pages by showing which URLs generate the most views over any time frame you specify. You can pinpoint exactly where your audience is spending their time and figure out what content’s actually working, without ever compromising user anonymity.
When you're analyzing site demographics, this tool gives you the whole picture. It breaks down traffic by URL so you know which pages are responsible for the most views. You get granular data on browser types and operating systems used across your entire user base. That means if you notice a spike in iOS activity but low performance metrics from Safari browsers, you'll know exactly where to focus your attention.
For understanding traffic sources or session health, the combined power of these tools lets you measure key stuff like average session duration for specific groups of users. You can query historical data and get ranked lists identifying major referral sources—you’ll see if a certain search engine or social platform is bringing in the most consistent traffic.
The whole point here is efficiency. Instead of asking your team to manually export CSVs, you just tell your agent what metric you need—active users, top URLs, device breakdown—and it spits clean, actionable numbers back. It’s all about getting reliable data without dealing with the complexity or tracking baggage that other platforms force on you.
How Umami Cloud MCP Works
- 1 First, you prompt your AI agent with a question—for example, 'Show me the top 5 pages by views last week.'
- 2 The agent recognizes the need for analytics data and invokes the Umami Cloud MCP Server tools (like
websites). - 3 The server executes the API call, fetches the raw metrics, and hands you the structured answer right back in your chat or code environment.
The bottom line is: it turns complex dashboard queries into simple conversational commands.
Who Is Umami Cloud MCP For?
Product Managers who need to justify feature decisions with hard usage data. Digital Marketers tracking campaign performance without worrying about cookie compliance. Web Analysts needing a quick, reliable way to pull segmented metrics that usually require hours in Google Sheets.
Using users and websites to test hypotheses: 'If we move the pricing button, will our active user count increase?' They need clean data fast.
Checking referral traffic sources or analyzing device usage via websites metrics before launching a paid campaign to ensure targeting is accurate.
Running comparative reports, like checking the bounce rate on mobile vs. desktop for the checkout page, all without leaving their terminal or IDE.
What Changes When You Connect
- Stop cross-referencing tabs. You get an exact count of active users with the
userstool—no more guessing how many people are on site right now. It's instant data. - Understand exactly who is visiting via
websites. You can break down traffic by OS, browser, and device type to spot user friction points immediately. - Analyze referral sources quickly. Instead of building custom reports, the agent uses the Umami API to list top referrers matching specific criteria.
- Assess site health with minimal effort. Calculate session metrics like bounce rates for a single URL on mobile devices without opening any dashboard.
- Keep your users safe while getting data. Because this server is built on Umami, you get deep analytics insights that don't rely on tracking cookies or PII.
Real-World Use Cases
Debugging a Mobile Drop-off
A PM notices conversion rates dip sharply after the checkout page. They ask their agent to run websites metrics for that URL, filtering by 'mobile' and comparing session duration against desktop usage. The agent returns the data: mobile users leave 30 seconds sooner than expected, pinpointing a potential UX issue.
Pre-Launch Campaign Validation
A Marketing Manager is about to launch a big social campaign. They ask their agent to list top referral sources for the last month. The agent uses this data to confirm that LinkedIn was, in fact, the highest contributor of quality traffic, allowing the manager to allocate budget correctly.
Quick Status Check
A developer needs a real-time headcount before starting work on an integration. They simply prompt for 'active users.' The agent runs users and gives them the immediate number, saving them from logging into a dashboard just for a count.
Comparing Site Performance
A Product Manager wants to know if their new landing page is performing better than the old one. They prompt the agent to pull websites metrics comparing /new-page vs. /old-page, getting a clean side-by-side view of unique visitors and device usage.
The Tradeoffs
Asking for everything at once
Prompting: 'Give me all the metrics, tell me about users, websites, bounce rate, and referral sources.' This overwhelms the agent and leads to an unmanageable data dump.
→
Break it down. First, run users for a simple headcount. Then, in a second prompt, focus: 'Now compare the websites metrics between desktop and mobile users on this specific URL.' Keep your requests surgical.
Assuming full data access
Asking an agent to pull historical user IDs or personally identifiable information (PII). The server will fail because Umami is privacy-focused.
→
Only ask for aggregated metrics. Stick to general counts, percentages, and device breakdowns—that's what the websites tool handles.
Manually scraping dashboards
Trying to copy and paste data from Umami's web interface into a spreadsheet when you only need a quick comparison.
→ Use the agent. Let your AI client run the query through the MCP Server, getting structured JSON output directly in your terminal or code.
When It Fits, When It Doesn't
Use this Umami Cloud server if your priority is gathering reliable web performance data while strictly maintaining user privacy. It’s perfect for developers and analysts who need to pull specific counts (users) or segment metrics (websites) on demand, without relying on third-party cookie tracking.
Don't use it if you need highly granular, individual user journey mapping—that requires a dedicated session recording tool. Also, don't use it if your primary goal is integrating deep CRM data; this server only handles web traffic metrics. If you just want to know who signed up (email addresses), this isn't the right tool; you need an API connected directly to your user database.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Umami Cloud. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 2 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Pulling website stats shouldn't require logging into five different dashboards.
Today, getting a clear picture of how users are behaving means clicking through Google Analytics, then switching tabs for device breakdowns, then exporting the page views to Excel. It’s tedious, it takes fifteen minutes, and you're probably doing it at 2 AM.
With Umami Cloud MCP Server, your agent runs the query for you. You ask: 'What are the top pages by view?' And in seconds, you get a clean list of URLs and counts right where you are working. Period.
Umami Cloud MCP Server: Get Clean Metrics on User & Website Performance
Manual data collection means losing context when copying numbers, or spending hours trying to reconcile which dashboard shows 'unique' vs. 'total' users. The risk of human error is huge.
Now, the agent handles the API calls and structures the output perfectly for your client. You get reliable, clean metrics—like an exact active user count via `users`—without touching a single dashboard interface.
Common Questions About Umami Cloud MCP
How does Umami Cloud MCP Server maintain privacy? +
It uses the Umami API, which is designed to be privacy-focused. It provides aggregated metrics like counts and percentages without tracking or exposing personal identifying information (PII) about individual users.
What can I get using the `users` tool? +
The users tool gives you one thing: a live count of how many people are currently active on your website. It's a simple, immediate headcount for real-time status checks.
Does Umami Cloud MCP Server handle historical data? +
Yes. You can prompt the agent to retrieve and analyze historical data, such as listing top pages from the last 7 days or analyzing monthly referral sources using websites.
Which metrics does the `websites` tool provide? +
The websites tool gives a detailed breakdown. You can analyze traffic by URL, browser type (Chrome vs Safari), operating system (iOS vs Windows), and device category (mobile/desktop).
Can I use Umami Cloud MCP Server with my code? +
Absolutely. Connecting to your agent via IDEs or terminal commands lets you run these analytics queries as part of a larger automated workflow, treating data retrieval like any other function call.
How do I handle authentication when using the `users` tool? +
You must provide a valid API key within your agent's context. The server validates this key before executing any calls to retrieve active user counts.
Are there rate limits if I query `websites` too frequently? +
Yes, the Umami Analytics API enforces standard request throttling rules. We recommend batching multiple metric requests into a single call to prevent exceeding defined usage limits.
What parameters can I pass to the `users` tool for advanced filtering? +
The users tool accepts optional filters, including geographic region and specific operating system types. This lets you narrow down active user counts beyond just a date range.
What data can AI generate from Umami Cloud? +
AI can query active users individually, metrics by website, page view breakdowns, browser/OS statistics, and custom triggered events.
How does the MCP authenticate with Umami Cloud? +
The integration uses an API access token. Provide your Umami Cloud token manually or pass username and password.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Apex Legends
Track Apex Legends player stats, match history, leaderboards, and server status directly from your AI agent.
Blastscan (Blast Network Explorer)
Explore the Blast Network—check balances, track transactions, and inspect smart contracts directly from your AI agent.
Dotloop
Manage real estate transactions via Dotloop — list loops, manage participants, and oversee documents.
You might also like
Treinta App
Empower Latin American SMBs automating Treinta app. Access ledgers, manage client debt, and review store inventories actively from chat.
Trimble
Equip your AI agent with enterprise-grade truck routing, commercial geocoding, and advanced toll cost logistics algorithms.
K-Fold Split Engine
Generate rigorous, leak-proof cross-validation indices for train and test splits in machine learning pipelines.