Pirsch Analytics MCP Server

Pirsch Analytics MCP Server

Enables natural language analytics queries for Pirsch Analytics, providing comprehensive website traffic statistics, period comparisons, and trend analysis through an MCP interface.

Category
Visit Server

README

MCP Pirsch Server

Version License

A Model Context Protocol (MCP) server for Pirsch Analytics, enabling natural language analytics queries, period comparisons, and trend analysis for your website traffic.

Features

  • ๐Ÿ” Smart Authentication - OAuth client credentials with automatic token caching and refresh
  • ๐Ÿ“Š Core Analytics - Comprehensive stats including visitors, page views, bounce rates, and conversion rates
  • ๐Ÿ“ˆ Time Series Data - Flexible visitor trends with day/week/month/year granularity
  • ๐Ÿ”„ Period Comparisons - Compare metrics across different time periods with calculated deltas
  • ๐ŸŽฏ Goals & Events - Read conversion goals, event activity, and event-specific page performance
  • ๐Ÿงญ Session Drilldown - Inspect entry pages, exit pages, session lists, and per-session timelines
  • โšก Real-time Insights - Active visitor tracking with configurable time windows
  • ๐ŸŽฏ Advanced Filtering - Full support for Pirsch query parameters including UTM, referrers, and dimensions
  • ๐ŸŒ Multi-domain Support - Manage analytics across multiple websites from one interface

Quick Start

Installation Methods

Option 1: Using NPX (No Installation Required)

The simplest way - no need to install anything globally:

# For Claude Desktop
npx @verygoodplugins/mcp-pirsch

# For Claude Code
claude mcp add pirsch "npx @verygoodplugins/mcp-pirsch"

Option 2: Global Installation

Install once, use anywhere:

# Install globally
npm install -g @verygoodplugins/mcp-pirsch

# For Claude Code
claude mcp add pirsch "mcp-pirsch"

Option 3: Local Development

For contributing or customization:

# Clone and install
git clone https://github.com/verygoodplugins/mcp-pirsch.git
cd mcp-pirsch
npm install
npm run build

Configuration

1. Get Pirsch API Credentials

  1. Log into your Pirsch Analytics Dashboard
  2. Navigate to Settings โ†’ API Clients
  3. Create a new client with appropriate permissions
  4. Copy your Client ID and Client Secret

2. Configure Your Client

<details> <summary><b>Claude Desktop Configuration</b></summary>

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "pirsch": {
      "command": "npx",
      "args": ["@verygoodplugins/mcp-pirsch"],
      "env": {
        "PIRSCH_CLIENT_ID": "your_client_id",
        "PIRSCH_CLIENT_SECRET": "your_client_secret",
        "PIRSCH_DEFAULT_DOMAIN_ID": "your_domain_id",
        "PIRSCH_TIMEZONE": "America/New_York"
      }
    }
  }
}

Or if installed globally:

{
  "mcpServers": {
    "pirsch": {
      "command": "mcp-pirsch",
      "env": {
        "PIRSCH_CLIENT_ID": "your_client_id",
        "PIRSCH_CLIENT_SECRET": "your_client_secret"
      }
    }
  }
}

</details>

<details> <summary><b>Claude Code Configuration</b></summary>

claude mcp add pirsch "npx @verygoodplugins/mcp-pirsch" \
  --env PIRSCH_CLIENT_ID=your_client_id \
  --env PIRSCH_CLIENT_SECRET=your_client_secret \
  --env PIRSCH_DEFAULT_DOMAIN_ID=your_domain_id

</details>

<details> <summary><b>Cursor IDE Configuration</b></summary>

Add to .mcp.json in your project:

{
  "mcpServers": {
    "pirsch": {
      "command": "node",
      "args": ["./node_modules/@verygoodplugins/mcp-pirsch/dist/index.js"],
      "env": {
        "PIRSCH_CLIENT_ID": "your_client_id",
        "PIRSCH_CLIENT_SECRET": "your_client_secret"
      }
    }
  }
}

</details>

3. Environment Variables

Create a .env file for local development:

# Required
PIRSCH_CLIENT_ID=your_client_id
PIRSCH_CLIENT_SECRET=your_client_secret

# Optional
PIRSCH_DEFAULT_DOMAIN_ID=your_domain_id  # Auto-detected if not set
PIRSCH_TIMEZONE=America/New_York         # Default: UTC
PIRSCH_TOKEN_SKEW_MS=60000              # Token refresh buffer (default: 60 seconds)

Available Tools

Discovery & Setup

pirsch_list_domains

List all accessible domains to discover domain IDs.

Parameters:

  • search (optional): Filter domains by name

Example:

List all my Pirsch domains

Core Statistics

pirsch_overview

Get cached overview statistics for a domain.

Parameters:

  • domain_id (optional): Target domain ID

Returns: Visitors, page views, and member counts

Note: This is the Pirsch cached overview endpoint. Filters do not apply, and it should not be used as a substitute for pirsch_total over a custom date range.

pirsch_total

Get total metrics for a specific period.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (optional): Filter object with date range, dimensions, etc.

Most analytics tools accept filter fields either inside filter or as top-level arguments. Both forms are supported for MCP client compatibility.

Returns: Total visitors, views, sessions, bounces, bounce rate, conversion rate, and custom metric aggregates

pirsch_visitors

Get visitor time series data.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (optional): Including scale (day/week/month/year)

Example:

Show me daily visitor trends for the last month

pirsch_pages

Get top pages with performance metrics.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (optional): Including:
    • sort: Sort field
    • direction: asc/desc
    • search: Search in page paths
    • include_avg_time_on_page: Include time metrics
    • include_title: Include page titles

Tip: Exact path: "/news/" still matches only that URL. For section queries on page-style tools, path-shaped values such as search: "/news/", path: "~/news/", or pattern: "/news/*" are narrowed again inside the MCP so /documentation/news/... does not leak into /news/... results. path_prefix is also available when you want an explicit root-prefix filter. Top-level search, path, and path_prefix arguments are also accepted.

pirsch_entry_pages

Get entry page analytics.

pirsch_exit_pages

Get exit page analytics.

pirsch_referrers

Analyze traffic sources and referrers.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (optional): Standard filter parameters

pirsch_goals

Get conversion goals together with their observed stats.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (optional): Standard filter parameters

pirsch_events

Get aggregated event statistics including counts, visitors, conversion rate, and metadata keys.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (optional): Standard filter parameters

pirsch_event_pages

Get pages on which a specific event fired.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (required): Standard filter parameters, including event

The event can also be passed as a top-level event argument. If your client uses goal payload field names, event_name is accepted as an alias and normalized to event. The same path-prefix narrowing described for pirsch_pages also applies here.

pirsch_utm

Analyze UTM campaign parameters.

Parameters:

  • type (required): source | medium | campaign | content | term
  • domain_id (optional): Target domain ID
  • filter (optional): Standard filter parameters

Example:

Show me UTM source breakdown for this week

pirsch_growth

Calculate growth rates across metrics.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (optional): Date range for growth calculation

Real-time Analytics

pirsch_active

Get currently active visitors and pages.

Parameters:

  • domain_id (optional): Target domain ID
  • start (optional): Seconds to look back (default: 600)

Example:

Show me active visitors in the last 5 minutes

Session Analytics

pirsch_sessions

Get session list data including entry/exit pages, duration, geography, device, and traffic source context.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (optional): Standard filter parameters

pirsch_session_details

Get the chronological page-view and event timeline for a single session.

Parameters:

  • domain_id (optional): Target domain ID
  • filter (required): Must include both visitor_id and session_id

Comparative Analytics

pirsch_compare

Compare metrics between two time periods using true period totals from Pirsch totals, plus the matching visitor series for charts.

Parameters:

  • domain_id (optional): Target domain ID
  • period (optional): today | yesterday | week | lastWeek | month | lastMonth
  • compare (optional): previous | year | custom
  • from, to (optional): Custom date range (YYYY-MM-DD)
  • compare_from, compare_to (optional): Custom comparison range
  • scale (optional): day | week | month | year

Example:

Compare this week's traffic to last week

Filter Parameters

Most tools accept a filter object that maps to Pirsch query parameters:

{
  // Date/Time
  "from": "2024-01-01",        // Start date (YYYY-MM-DD)
  "to": "2024-01-31",          // End date (YYYY-MM-DD)
  "from_time": "09:00",        // Start time (HH:MM)
  "to_time": "17:00",          // End time (HH:MM)
  "tz": "America/New_York",    // Timezone
  
  // Dimensions
  "path": "~/news/",           // Exact or operator-based path filter (~ contains, ! not, ^ does-not-contain)
  "entry_path": "/landing",    // Entry page
  "exit_path": "/checkout",    // Exit page
  "pattern": "*.pdf",          // URL pattern
  
  // Traffic Sources
  "referrer": "google.com",    // Referrer domain
  "referrer_name": "Google",   // Referrer name
  "channel": "organic",        // Traffic channel
  
  // UTM Parameters
  "utm_source": "newsletter",
  "utm_medium": "email",
  "utm_campaign": "summer-sale",
  "utm_content": "header-cta",
  "utm_term": "analytics",
  
  // Device/Browser
  "os": "Windows",
  "browser": "Chrome",
  "platform": "desktop",       // desktop | mobile | unknown
  "screen_class": "xxl",
  
  // Location
  "country": "US",
  "city": "New York",
  "language": "en",
  
  // Pagination/Sorting
  "offset": 0,
  "limit": 100,
  "sort": "visitors",
  "direction": "desc",         // asc | desc
  "search": "/news/",          // Path-shaped searches are narrowed to root-prefix matches on page-style tools
  "path_prefix": "/news/",     // Optional explicit MCP-local prefix matcher for page-style tools
  "keyword": "wordpress crm",  // Google Search Console keyword filter
  
  // Advanced
  "event": "signup",
  "event_meta_key": "plan",
  "tag": "premium",
  "visitor_id": "12345...",    // Required together with session_id for pirsch_session_details
  "session_id": "67890",
  "custom_metric_key": "revenue",
  "custom_metric_type": "float"
}

Usage Examples

Basic Analytics Query

Show me the visitor statistics for last week

Page Performance Analysis

What are my top 10 /news/ posts by traffic this month?

Campaign Tracking

Analyze UTM campaign performance for the summer sale

Traffic Sources

Show me referrer breakdown excluding direct traffic

Period Comparison

Compare this month's metrics to the same period last year

Real-time Monitoring

How many people are on my site right now?

Goals and Events

Show me conversion goals and top event-driven pages for the last 90 days

Session Investigation

List recent sessions that entered on /news/ and inspect one session in detail

Development

Building from Source

npm install
npm run build

Development Mode

npm run dev  # Watch mode with auto-reload

Testing

npm test

Troubleshooting

Authentication Issues

Invalid credentials error

  • Verify your Client ID and Secret are correct
  • Check that your API client has appropriate permissions in Pirsch
  • Ensure credentials are properly set in environment variables

Token refresh failures

  • The server automatically refreshes tokens 60 seconds before expiry
  • Check network connectivity to Pirsch API
  • Verify PIRSCH_TOKEN_SKEW_MS is not set too low

Domain Issues

Domain not found

  • Run pirsch_list_domains to see available domains
  • Verify PIRSCH_DEFAULT_DOMAIN_ID is correct
  • Check API client has access to the domain

No data returned

  • Verify the date range contains data
  • Check timezone settings match your Pirsch configuration
  • Ensure proper filtering parameters
  • Use pirsch_total for custom date range totals; pirsch_overview is cached and not filterable
  • For page-style tools, path-shaped search, ~/path/, and /path/* filters are narrowed to root-prefix matches inside the MCP
  • Use path_prefix when you want explicit prefix behavior without relying on Pirsch operators

Performance

Slow responses

  • Token caching reduces authentication overhead
  • Consider adjusting PIRSCH_TOKEN_SKEW_MS for your use case
  • Check network latency to Pirsch API endpoints

Contributing

Contributions are welcome! Please:

  1. Fork the repository
  2. Create a feature branch
  3. Make your changes with tests
  4. Submit a pull request

License

MIT - See LICENSE file for details.

Support

Credits

Built by Jack Arturo ๐Ÿงก

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured
E2B

E2B

Using MCP to run code via e2b.

Official
Featured