StatSource

Statsource MCP Server: AI-Powered Data Analysis & Statistical Insights

A standalone Model Context Protocol server that connects your PostgreSQL data and CSV files to AI-powered analysis and statistics – all in plain English.

Get Started with StatsourceSign In
statsource@localhost:~$ What's the average temperature in July?
Analysis Result

The average temperature in July is 28.5°C.

Based on data from weather_data.csv (10,000 entries)

What is Statsource?

Statsource is a standalone MCP (Model Context Protocol) server designed to simplify data analysis. Acting as a middleware, it connects clients like Claude Desktop to your existing API, enabling natural language queries and advanced statistical calculations.

Whether you're pulling data from a PostgreSQL database or a CSV file, Statsource delivers actionable insights with ease. Open-source and highly configurable, it's the perfect tool for unlocking the power of your data.

Available Tools

get_statistics

Analyze data and calculate statistics or generate ML predictions based on provided parameters.

Arguments:

  • columns (list of strings, required): List of column names to analyze or predict (Ask user for exact column names).
  • data_source (string, optional): Path to data file (uploaded to statsource.me), database connection string (ask user for exact string), or API endpoint. Defaults to `DB_CONNECTION_STRING` env var if set.
  • source_type (string, optional): Type of data source ("csv", "database", or "api"). Defaults to `DB_SOURCE_TYPE` env var if set.
  • table_name (string, optional but **required** if `source_type` is "database"): Name of the database table to use (Ask user for exact table name).
  • statistics (list of strings, optional): List of statistics to calculate (required for `query_type="statistics"`). Valid options include: 'mean', 'median', 'std', 'sum', 'count', 'min', 'max', 'describe', 'correlation', 'missing', 'unique', 'boxplot'.
  • query_type (string, optional, default="statistics"): Type of query ("statistics" or "ml_prediction").
  • periods (integer, optional): Number of future periods to predict (required for `query_type="ml_prediction"`).
  • filters (dict, optional): Dictionary of column-value pairs to filter data (e.g., `"status": "completed", "region": ["North", "East"]`).
  • groupby (list of strings, optional): List of column names to group data by before calculating statistics (e.g., `["region", "product_category"]`).
  • options (dict, optional): Dictionary of additional options for specific operations.
  • date_column (string, optional): Column name containing date/timestamp information for filtering and time-series analysis.
  • start_date (string or datetime, optional): Inclusive start date for filtering (ISO 8601 format, e.g., "2023-01-01").
  • end_date (string or datetime, optional): Inclusive end date for filtering (ISO 8601 format, e.g., "2023-12-31").

Note: For CSV, upload files to statsource.me first. For databases, always ask the user for the exact connection string and table name. Do not guess connection details. If `data_source` and `source_type` are omitted, environment variables (`DB_CONNECTION_STRING`, `DB_SOURCE_TYPE`) will be used if configured.

suggest_feature

Suggest a new feature or improvement for the StatSource analytics platform.

Arguments:

  • description (string, required): A clear, detailed description of the suggested feature
  • use_case (string, required): Explanation of how and why users would use this feature
  • priority (string, optional): Suggested priority level ("low", "medium", "high")

Why Statsource Stands Out

Natural Language Queries

Ask questions like 'What's the average sales in my food_sales table?' and get instant answers.

Statistical Analysis

Calculate means, medians, standard deviations, and more with built-in tools.

Flexible Data Sources

Works with databases, CSV files, or API data – your choice.

AI-Powered Insights

Predict trends or analyze data with advanced AI processing.

Analyze Data in Three Simple Steps

1

Connect Your Data

Link Statsource to your database, CSV, or API.

2

Ask a Question

Use plain English, like "Predict sales for next month."

3

Get Insights

Receive accurate statistics or predictions instantly.

Example

Query: "What will the next 5 periods of sales be?"
Result:

I'll use the get_statistics tool to predict the next 5 periods of sales from your data.

Forecast values:

  • Period 1: 191.02
  • Period 2: 191.38
  • Period 3: 191.74
  • Period 4: 192.10
  • Period 5: 192.46

Analysis:

  • Increasing trend direction
  • Strong model fit (R-squared: 0.9999)
  • Positive slope of 0.36 per period

These predictions show a steady upward trend in your sales. The high R-squared value indicates the model has found a highly predictable pattern in your historical data.

Unlock the Value of Your Data

Statsource saves time, reduces errors, and empowers data-driven decisions. Ideal for data analysts, developers, and businesses, it turns complex data into simple insights – no coding required.

Start Using Statsource Today

Ready to explore your data? Install the Statsource MCP server in minutes with our open-source package and connect it to your AI agent. Try it free and see the difference.

Installation with uv (Recommended)

Using uv requires no explicit installation step. Use `uvx` to run directly:

Configure for Claude.app:

"mcpServers": {
  "statsource": {
    "command": "uvx",
    "args": ["mcp-server-stats"]
  }
}

Installation with PIP

Alternatively, install using pip:

pip install mcp-server-stats

Run via module or console script:

python -m mcp_server_stats mcp-server-stats

Configure for Claude.app:

"mcpServers": {
  "statsource": {
    "command": "python",
    "args": ["-m", "mcp_server_stats"]
  }
}

Installation with Docker

Use the official Docker image:

Configure for Claude.app:

"mcpServers": {
  "statsource": {
    "command": "docker",
    "args": ["run", "-i", "--rm", "statsource/mcp"]
  }
}

Configuration via Environment Variables

Configure defaults and API keys in your Claude.app settings:

"mcpServers": {
  "statsource": {
    "command": "python", // or "uvx", "docker"
    "args": ["-m", "mcp_server_stats"], // Adjust based on command
    "env": {
      "API_KEY": "your_api_key",
      "DB_CONNECTION_STRING": "postgresql://user:pass@host:port/db",
      "DB_SOURCE_TYPE": "database"
    }
  }
}

Available Variables:

  • API_KEY: Your API key for statsource.me
  • DB_CONNECTION_STRING: Default database connection string
  • DB_SOURCE_TYPE: Default source type (e.g., "database")

Debugging

Use the MCP Inspector to debug:

npx @modelcontextprotocol/inspector uvx mcp-server-stats

Or, if installed locally:

npx @modelcontextprotocol/inspector python -m mcp_server_stats
Get Your API KeyView on GitHub