MCP Analytics Suite

MCP server for data analytics — Shopify, Stripe, WooCommerce, eBay, CSV files, and more. Run statistical analysis, forecasting, and machine learning directly in Claude or Cursor. Ask a question, upload your data, get an interactive report.

This is the public listing and documentation repository. Issues, feature requests, and examples live here. The API server code is maintained separately.

Sample Reports → • Try Demo → • Pricing →

Every analysis starts with a question. We handle the rest.

🚀 Quick Start • 🔄 How It Works • 🛠️ MCP Tools • 🛡️ Security • 📖 Documentation

The Formula

Question + Dataset = Analytics

Transform business questions into actionable insights through intelligent discovery

Overview

MCP Analytics Suite is an intelligent analytics platform that understands what you want to analyze and automatically selects the right approach. No statistics degree required — just describe your business question and let our AI-powered discovery handle the complexity.

Upload any CSV — Shopify orders, Stripe exports, WooCommerce reports, eBay data, ad platform reports, or any tabular data. Connect live data from Google Analytics 4 and Google Search Console via native connectors. Run regression, forecasting, clustering, A/B testing, customer LTV, churn prediction, and hundreds of other statistical methods. Get back interactive HTML reports with charts and AI-written insights.

Why MCP Analytics?

Intelligent Discovery: Automatically finds the right analytical approach
Complete Workflow: From question to insight in one seamless flow
Zero Setup: Cloud-based processing, works instantly
Enterprise Security: OAuth2, encryption, isolated processing
Comprehensive Suite: Full range of analytical capabilities
Interactive Reports: Shareable visualizations with AI insights

Quick Start

Installation

For Claude Desktop

Add to your config file:

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

{
  "mcpServers": {
    "mcp-analytics": {
      "command": "npx",
      "args": ["-y", "mcp-remote@latest", "https://api.mcpanalytics.ai/auth0"]
    }
  }
}

For Cursor

Add to .cursor/config.json in your project root:

{
  "mcpServers": {
    "mcp-analytics": {
      "command": "npx",
      "args": ["-y", "mcp-remote@latest", "https://api.mcpanalytics.ai/auth0"]
    }
  }
}

For VS Code (Continue Extension)

Add to your Continue config at ~/.continue/config.json:

{
  "models": [{
    "provider": "anthropic",
    "model": "claude-3-5-sonnet",
    "mcpServers": {
      "mcp-analytics": {
        "command": "npx",
        "args": ["-y", "mcp-remote@latest", "https://api.mcpanalytics.ai/auth0"]
      }
    }
  }]
}

For Claude Code

Add to claude_code_config.json:

{
  "mcpServers": {
    "mcp-analytics": {
      "command": "npx",
      "args": ["-y", "mcp-remote@latest", "https://api.mcpanalytics.ai/auth0"]
    }
  }
}

How It Works

The MCP Analytics Workflow

Ask Your Question - Describe what you want to analyze in natural language
Intelligent Discovery - tools.discover finds the right analytical approach
Data Upload - datasets.upload securely processes your data
Automated Analysis - tools.run executes with optimal configuration
Interactive Results - reports.view delivers shareable insights

User: "What drives our sales growth?"
MCP Analytics:
  → Discovers regression and correlation methods
  → Configures analysis for your data structure
  → Runs multiple analytical approaches
  → Returns comprehensive report with insights

MCP Tools

The platform provides a complete suite of MCP tools for end-to-end analytics:

Core Analytics Tools

discover_tools - Natural language tool discovery (5-signal semantic search)
tools_run - Execute an analysis module on your data
tools_info - Get tool documentation and schema
tools_schema - Inspect column requirements for a tool

Data Management

datasets_upload - Secure data upload with encryption
datasets_list - List your uploaded datasets
datasets_read - Preview dataset contents
datasets_download - Download a dataset
datasets_update - Update dataset metadata

Connectors

connectors_list - List available data source connections
connectors_query - Pull live data from a connected source

Reporting & Insights

reports_view - Open an interactive HTML report
reports_list - List your reports
reports_search - Semantic search across past analyses
agent_advisor - Conversational AI that guides analysis and interprets results

Platform Tools

billing - Usage and subscription management
about - Platform information and status

Features

Natural Language Interface

Just describe what you need:

"What drives our revenue growth?"
"Find customer segments in our data"
"Forecast next quarter's sales"
"Did our marketing campaign work?"

Comprehensive Analysis Suite

Statistical Methods

Regression Analysis
Advanced Modeling
Hypothesis Testing
Survival Analysis
Bayesian Methods

Machine Learning

Ensemble Methods
Boosting Algorithms
Neural Networks
Clustering
Dimensionality Reduction

Time Series

Forecasting
Seasonal Analysis
Trend Detection
Multivariate Models
Causal Analysis

Business Analytics

Customer Analytics
Market Analysis
Pricing Models
Predictive Analytics
Experimental Design

Seamless Workflow

graph LR
    A[Ask in Claude/Cursor] --> B[MCP Analytics]
    B --> C[Secure Processing]
    C --> D[Interactive Report]
    D --> E[Share Results]

Example Usage

Basic Regression

User: "I have a CSV with house prices. Can you predict price based on size and location?"
Claude: [Runs linear regression, provides R², coefficients, and diagnostic plots]

Customer Segmentation

User: "Segment my customers in sales_data.csv into meaningful groups"
Claude: [Performs k-means clustering, creates segment profiles with visualizations]

Time Series Forecasting

User: "Forecast next quarter's revenue using our historical data"
Claude: [Applies ARIMA, generates predictions with confidence intervals]

Security & Compliance

Enterprise Security Features

Authentication: OAuth2 via Auth0 with PKCE
Encryption: TLS 1.3 for all data transfers
Processing: Isolated Docker containers per analysis
Data Handling: Ephemeral processing, no persistence
Access Control: OAuth 2.0 scoped permissions with usage limits
Audit Trail: Complete logging for compliance

Privacy & Data Handling

Data Privacy: Ephemeral processing, no data retention
User Rights: Data deletion upon request
Secure Processing: Isolated containers per analysis
Enterprise Options: Contact us for compliance requirements

Read full security documentation →

Architecture

flowchart TB
    subgraph "Client Integration"
        CLI[CLI/SDK]
        Claude[Claude Desktop]
        Cursor[Cursor IDE]
        MCP[MCP Protocol]
    end

    subgraph "API Gateway"
        LB[Load Balancer]
        Auth[OAuth 2.0/Auth0]
        Rate[Rate Limiting]
    end

    subgraph "Processing Layer"
        Router[Request Router]
        Queue[Job Queue]
        Workers[Processing Workers]
        Docker[Docker Containers]
    end

    subgraph "Analytics Engine"
        Stats[Statistical Methods]
        ML[Machine Learning]
        TS[Time Series]
        Report[Report Generation]
    end

    subgraph "Data Layer"
        Cache[Results Cache]
        Storage[Secure Storage]
        Encrypt[Encryption Layer]
    end

    CLI --> LB
    Claude --> LB
    Cursor --> LB
    MCP --> LB

    LB --> Auth
    Auth --> Rate
    Rate --> Router

    Router --> Queue
    Queue --> Workers
    Workers --> Docker

    Docker --> Stats
    Docker --> ML
    Docker --> TS

    Stats --> Report
    ML --> Report
    TS --> Report

    Report --> Cache
    Cache --> Storage
    Storage --> Encrypt

    style Auth fill:#e8f5e9
    style Docker fill:#fff3e0
    style Report fill:#e3f2fd

Performance

Dataset Size: Handles large datasets
Processing Time: Fast cloud-based processing
Secure Infrastructure: Isolated Docker containers
API Access: RESTful API with authentication

Getting Started

Visit our website for pricing and signup →

Documentation

Quick Start Guide - Get running in under a minute
Architecture - How the platform works
Connectors - GA4, GSC, and CSV data sources
Pricing - Plans and limits
Security - Security & compliance details
API Reference - Complete API documentation
Tutorials - Step-by-step guides

Support

Issues: GitHub Issues
Email: support@mcpanalytics.ai
Docs: mcpanalytics.ai/docs
Enterprise: sales@mcpanalytics.ai

Comparison with Other MCP Servers

Feature	MCP Analytics	Google Analytics MCP	PostgreSQL MCP	Filesystem MCP
Use Case	Statistical Analysis	Web Metrics	Database Queries	File Access
Setup Time	30 seconds	OAuth + Config	Connection string	Path config
Data Sources	Any CSV/JSON/URL	GA4 Only	PostgreSQL Only	Local files
Analysis Tools	Full Suite	GA4 Metrics	SQL Only	Read/Write
Machine Learning	✅ Full Suite	❌	❌	❌
Visualizations	✅ Interactive	✅ Dashboards	❌	❌
Shareable Reports	✅	❌	❌	❌

Detailed comparison →

About MCP Analytics

MCP Analytics is built by data scientists and engineers passionate about making advanced statistical analysis accessible through AI assistants. The platform runs validated, deterministic analysis modules — the same data and tool produce the same result every time, unlike LLM code generation.

Testing & Support

Testing Your Connection

After installation, restart your IDE and look for "MCP Analytics" in the available tools. On first use, you'll be prompted to authenticate via OAuth 2.0.

# To test the connection directly:
npx -y mcp-remote@latest https://api.mcpanalytics.ai/auth0

Troubleshooting

If MCP Analytics doesn't appear after installation:

Ensure your config file is valid JSON
Restart your IDE completely
Check the IDE's developer console for errors
Verify you have internet connectivity

For support: support@mcpanalytics.ai

Contributing

While the core server is proprietary, we welcome contributions to:

Documentation improvements
Example notebooks and use cases
Bug reports and feature requests
Community tools and integrations

See CONTRIBUTING.md for guidelines.

License

MCP Analytics is a product of PeopleDrivenAI LLC.

This is commercial software. Use of the MCP Analytics service is subject to our:

Ready to transform your data analysis workflow?

Get Started Free | Read Docs | View Demo

Built by MCP Analytics | Powered by R & Python

If MCP Analytics saves you time, a ⭐ on GitHub helps others find it.

Tags: mcp mcp-server model-context-protocol analytics data-analytics shopify-analytics stripe-analytics csv-analysis statistics machine-learning time-series clustering regression business-intelligence claude cursor ai-tools no-code-analytics forecasting customer-analytics

Name		Name	Last commit message	Last commit date
Latest commit History 10 Commits
.claude-plugin		.claude-plugin
.github		.github
docs		docs
examples		examples
.gitignore		.gitignore
CHANGELOG.md		CHANGELOG.md
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
CONTRIBUTING.md		CONTRIBUTING.md
LICENSE		LICENSE
README.md		README.md
SECURITY.md		SECURITY.md
server.json		server.json

Folders and files

Latest commit

History

Repository files navigation

MCP Analytics Suite

The Formula

Question + Dataset = Analytics

Overview

Why MCP Analytics?

Quick Start

Installation

For Claude Desktop

For Cursor

For VS Code (Continue Extension)

For Claude Code

How It Works

The MCP Analytics Workflow

MCP Tools

Core Analytics Tools

Data Management

Connectors

Reporting & Insights

Platform Tools

Features

Natural Language Interface

Comprehensive Analysis Suite

Seamless Workflow

Example Usage

Basic Regression

Customer Segmentation

Time Series Forecasting

Security & Compliance

Enterprise Security Features

Privacy & Data Handling

Architecture

Performance

Getting Started

Documentation

Support

Comparison with Other MCP Servers

About MCP Analytics

Testing & Support

Testing Your Connection

Troubleshooting

Contributing

License

About

Topics

Resources

License

Code of conduct

Contributing

Security policy

Uh oh!

Stars

Watchers

Forks

Releases 1

Contributors

Uh oh!