Supercharge your AI with Pirsch Analytics. Query web traffic and user behavior metrics instantly.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Connect to your AI in seconds.
Pirsch Analytics connects your AI client directly to web analytics data. It lets you query site performance—visitor counts, event completions, page views, and traffic sources—using natural language commands instead of opening a dashboard.
You can also programmatically manage domains and send real-time tracking hits from within your agent workflow.
What your AI can do
Create domain
Adds a new website domain to the Pirsch analytics system.
List domains
Returns a list of every domain currently configured in your Pirsch account.
Send event batch
Submits multiple user events in one call for efficient bulk tracking.
List all connected domains or create a new one directly through an API call.
Retrieve structured statistics on site visits, active users, page performance, and goal completions for specific time periods.
Stream web activity data by sending single hits (send_hit) or batches of events/hits to log user actions in real-time.
Ask an AI about this
Compatible AI Apps
OAuth 2.0 Compatible Gemini Waiting for input…
Pirsch Analytics MCP Server: 14 Tools for Web Data Management
These tools let you manage domains, send real-time hits and events, and query every type of site metric—from unique visitors to campaign sources.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Pirsch Analytics on VinkiusCreate Domain
Adds a new website domain to the Pirsch analytics system.
List Domains
Returns a list of every domain currently configured in your Pirsch account.
Send Event Batch
Submits multiple user events in one call for efficient bulk tracking.
Send Event
Sends a single, specific user event (like 'button_click') to be tracked by Pirsch.
Send Hit Batch
Submits multiple page views (hits) in one call to log several site visits...
Send Hit
Tracks a single page view (a 'hit') and sends as much information as possible for accurate analytics.
Get Statistics Active
Retrieves statistics showing the count of currently active visitors on a domain.
Get Statistics Events
Gets a list of all tracked user events and their respective counts.
Get Statistics Goals
Pulls statistics specifically related to defined conversion goals on the site.
Get Statistics Overview
Retrieves a high-level summary of all key website metrics for a specified domain and...
Get Statistics Page
Gets detailed statistics about specific pages, including view counts and traffic...
Get Statistics Referrer
Analyzes where the user came from by listing detailed referrer data (e.g., google.com/search).
Get Statistics Utm Source
Gathers metrics detailing traffic sources based on UTM parameters for campaign tracking.
Get Statistics Visitor
Retrieves historical data and counts for unique visitors to the domain over time.
Connect to your AI in seconds. Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token from
cloud.vinkius.com. For OAuth-protected servers, expand Advanced
settings to add credentials.
Start a conversation
Open a new chat. The Pirsch Analytics integration is available immediately — no restart needed.
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 Pirsch Analytics, then connect any of our 5,000+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,000+ 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
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Pirsch Analytics. 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 connection provides 14 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Sifting through dashboards to find one simple metric shouldn't take five minutes.
Today, if you want to know how many people visited your checkout page last week, you have to open the analytics dashboard. Then, you select the date range, navigate to 'Pages,' filter by the URL path, and finally, download a CSV just to find the total count. It's clicking, filtering, and copy-pasting.
With Pirsch Analytics connected, you simply ask: 'What were the page view statistics for /checkout last week?' The agent runs `get_statistics_page`, pulls the exact metric, formats it cleanly, and gives you the number instantly. No dashboard needed.
Using send_hit and send_event_batch: Programmatic tracking is a game-changer.
Manually setting up test user flows or simulating complex interactions for testing used to mean writing messy, dedicated scripts that only logged basic data. You had to manually manage the payload structure every time you needed to test a new conversion step.
Now, your agent handles it. You use `send_hit` for page views and `send_event_batch` for structured user actions. Your workflow stays clean because the server manages the complex analytics data modeling behind the scenes.
What your AI can actually do with this
Yo, peep this: Pirsch Analytics hooks up your AI client straight to web analytics data. You can query site performance—visitor counts, event completions, page views, and traffic sources—using plain English commands instead of having to open some clunky dashboard. Plus, you can programmatically manage domains and send real-time tracking hits right from within your agent's workflow.
Domain Management
You can keep track of all the sites you're playing with using list_domains, which gives you a list of every domain already set up in your Pirsch account. If you need to start monitoring a new site, just run create_domain to add it to the system.
Traffic & Behavior Querying
When it comes to digging into stats, this server lets you pull structured metrics on everything from active users to goal completions for specific time frames. You can get a high-level view of all key website numbers—total visitors, page views, etc.—using get_statistics_overview for any domain and date range. Need more detail? Use get_statistics_page to pull stats on specific pages, including their traffic volume and view counts.
For active users, get_statistics_active shows you how many people are currently hitting the site. To track conversions, get_statistics_goals pulls statistics specifically related to your defined conversion goals. You can also see where your visitors came from by running get_statistics_referrer, which gives detailed data on the referring site (like google.com/search). If you need to track marketing campaigns, get_statistics_utm_source gathers metrics based on UTM parameters so you know exactly what worked.
Over time, you can analyze historical visitor counts using get_statistics_visitor.
Data Ingestion: Sending Web Activity Data
This is how you feed data back into the system in real-time. You track a single page view (a 'hit') by calling send_hit, which sends as much info as possible to keep your analytics accurate. If you're logging multiple user actions, you don't want to spam the API; use send_event_batch or send_hit_batch to submit several events or hits in one efficient call.
You can also send a single, specific user event—like a button click—using send_event. It’s all about making sure your agent logs every move.
Basically, you tell your AI client what data you need—say, 'Show me the total visitors for my e-commerce site last quarter,' or 'Log this purchase event now.' The agent picks and runs the right tool call and spits back structured metrics immediately. You're not stuck staring at a dashboard; your agent handles it all.
019e38d7-52ce-705a-a73e-10b3696915e7 
Here's how it actually works
The bottom line is that your AI client executes these analytics tools as if they were functions in Python, giving you real-time data access without needing to open any web dashboard.
Subscribe to the server and provide your Pirsch Access Token (Client Secret) to the agent.
Tell your AI client the exact data point you need. Example: 'What were the visitor counts for domain ABC in November?'
The agent calls the appropriate tool (get_statistics_visitor, etc.) using structured arguments, and you get a clean JSON response with the requested metrics.
Who is this actually for?
This is for the developer or data analyst who spends too much time clicking through multiple dashboards just to find one key metric. If your job involves tracking site performance, campaign metrics, or auditing user flows, this server gets you raw data instantly.
Needs to extract raw statistics (like visitor counts or UTM sources) for custom reporting and audit purposes without manual export/cleaning.
Needs instant campaign performance reports, such as goal completion metrics or unique visitor trends, directly in a chat window instead of navigating to the marketing dashboard.
Checks site health and logs test user flows by programmatically sending hits (send_hit) or validating domain configurations using list_domains or create_domain.
What Changes When You Connect
Get immediate insights on visitor patterns. Instead of clicking through multiple tabs to check stats, use get_statistics_overview to pull total counts, unique visitors (get_statistics_visitor), and page views in a single query.
Track user actions programmatically. Send hits or events using tools like send_hit or send_event_batch. This lets your agent simulate traffic flow for testing or record complex interactions without manual intervention.
Pinpoint campaign success immediately. When you need to know if a marketing push worked, call get_statistics_utm_source. It pulls detailed source data so you don't have to dig through spreadsheet columns.
Manage your analytics stack from the chat. Need to check or add a domain? Use list_domains or create_domain to keep your infrastructure management right where your code is running, eliminating context switching.
Understand traffic flow deeply. Don't just know that people came; use get_statistics_referrer and get_statistics_page together to see exactly which source drove the user to a specific page.
See it in action
Auditing Site Health Post-Deployment
A developer needs to validate that all new tracking is working. They first run list_domains to get the target ID, then use send_hit with a test URL and IP address. Finally, they call get_statistics_page to confirm the hit was registered correctly.
Analyzing Campaign Attribution
A marketing analyst runs a campaign from LinkedIn. They ask their agent: 'Show me the visitor stats for domain X using only UTM sources that match LinkedIn.' The agent uses get_statistics_utm_source to isolate the specific performance data.
Monitoring Conversion Funnels
A product manager wants to track if users complete a complex sign-up sequence. They use send_event_batch to log 'step1_complete', then 'step2_submit'. Later, they call get_statistics_goals to see the conversion rate.
Quarterly Performance Review
A data analyst needs a raw report on all site activity for Q3. They use get_statistics_overview, specifying the start and end dates, which gathers visitor counts, page views, and goal metrics in one structured call.
The honest tradeoffs
Treating it like a simple logging tool
Just sending raw data without structure. You run send_hit but forget to specify the page path, so you get generic activity logs that don't show which section of the site was viewed.
Always use the specific statistical retrieval tools (get_statistics_page, get_statistics_overview) after sending data. These functions interpret the raw hits and give you actionable metrics, not just a log dump.
Mixing domain IDs and URLs
Telling the agent to 'Get stats for my site' without specifying which domain ID it should pull from. The agent gets confused about which dataset to query.
Always start by confirming your domains using list_domains. Then, when querying statistics, explicitly pass the correct Domain ID into tools like get_statistics_visitor.
Over-relying on single event tracking
Only calling send_event for button clicks. This gives you no idea if those clicks are happening because a user found the page or because they came from a specific campaign source.
Combine tools: first, use get_statistics_utm_source to understand traffic origin, and then pair it with send_event to track the resulting interaction.
When It Fits, When It Doesn't
Use this MCP Server if your primary need is structured access to historical web metrics. You want to run queries like 'What were my unique visitor counts last month, broken down by source?'—that's exactly what it does. It excels at data retrieval and programmatic event logging.
Don't use it if you only need a simple log dump or general server monitoring (like CPU usage). For those tasks, look into specialized infrastructure monitoring tools. Also, don't rely on it for real-time database writes; it handles the analytics layer above that. If your data schema is non-standard and requires heavy pre-processing before analysis, you might need a custom backend service instead of just querying these 14 predefined endpoints.
Questions you might have
How do I check my current traffic stats using get_statistics_overview? +
You provide a domain ID and specify the date range. The tool returns one comprehensive view containing key metrics like total hits, active visitors, and page views all in one response.
What's the difference between send_hit and send_event? +
A send_hit tracks a basic page load (a visit). A send_event tracks a specific user action, like clicking 'Submit' or viewing a video. You use both to build a full picture.
Can I list domains before sending data? +
Yes. Use the list_domains tool first. This confirms all available domain IDs are correctly registered with Pirsch before you try to send any tracking hits or events.
How do I get campaign source details using get_statistics_utm_source? +
Just ask for the UTM source statistics, providing the relevant date range. This function pulls detailed data showing exactly which marketing channel (e.g., Facebook, Google) drove the traffic.
What should I do if my API token expires when I run `list_domains`? +
The agent will receive a 401 Unauthorized error. You need to refresh your Pirsch Access Token via the main dashboard and update the client secret in Vinkius before retrying the query.
Is using `send_event_batch` better than calling `send_event` multiple times? +
Yes, always use batch actions for efficiency. Sending a large group of events through send_event_batch minimizes API calls and drastically improves the speed at which your data gets processed.
Do I have to call `create_domain` before I can get any statistics? +
Yep, you must run create_domain first. The server needs to know about the domain ID before it can fetch anything using tools like get_statistics_overview or get_statistics_page.
How does `get_statistics_goals` differ from simply checking visitor stats? +
Goals track specific, defined conversions—like sign-ups or purchases. While visitor data tells you who came, the goals tool confirms if they actually completed a critical action.
How can I get a quick summary of my website's performance for a specific period? +
You can use the get_statistics_overview tool. Just provide the Domain ID and the start/end dates (YYYY-MM-DD). The agent will return key metrics like visitors, views, and bounce rate.
Is it possible to track custom user actions like button clicks? +
Yes! Use the send_event tool. You'll need to provide the event name, the current URL, and visitor details (IP and User-Agent) to record the interaction in Pirsch.
Can I see which pages are the most popular on my site? +
Absolutely. Use the get_statistics_page tool with your Domain ID and date range. It will list the top pages along with their respective view counts and visitor data.
We've already built the connector for Pirsch Analytics. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 14 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.