Excel to Polars MCP Server

A Model Context Protocol (MCP) server that converts Excel files to Polars DataFrames using FastMCP. Includes comprehensive actuarial data examples and advanced analytics capabilities.

Features

• Convert Excel files (.xlsx, .xls) to Polars DataFrames
• Support for multiple sheets with automatic detection
• Configurable data type inference and schema validation
• Memory-efficient processing using Polars
• Multiple output formats: Parquet, CSV, JSON
• Schema generation with statistics and metadata
• Professional actuarial data examples
• Advanced Polars analytics and reporting
• Comprehensive test suite with realistic data

Installation

This project uses uv for dependency management. Make sure you have uv installed:

curl -LsSf https://astral.sh/uv/install.sh | sh

Then install the project:

uv sync

Usage

Running the MCP Server

uv run excel-polars-mcp

Available MCP Tools

• read_excel: Convert an Excel file to Polars DataFrame with configurable options
• list_sheets: List all sheet names in an Excel file
• read_excel_sheet: Read a specific sheet from an Excel file

API Usage

import polars as pl

# Load converted actuarial data
life_table = pl.read_parquet('output/life_table.parquet')
policies = pl.read_parquet('output/policies.parquet')
claims = pl.read_parquet('output/claims.parquet')
reserves = pl.read_parquet('output/reserves.parquet')

# Perform actuarial analysis
loss_ratios = policies.join(claims, on='Policy_ID').group_by('Policy_Type').agg([
    pl.col('Claim_Amount').sum() / pl.col('Face_Amount').sum() * 100
])

# High-value claims analysis
high_value_claims = claims.filter(pl.col('Claim_Amount') > 400000)

# Mortality analysis
high_mortality = life_table.filter(pl.col('Mortality_Rate_qx') > 0.1)

Output Formats

Each Excel sheet is converted to multiple formats:

Sample Data

The project includes comprehensive actuarial datasets:

Examples

Basic Usage

python3 examples/demo.py

Actuarial Data Example

A comprehensive example that creates sample actuarial data and converts it to Polars format:

python3 run_actuarial_example.py

This example:

1. Generates a multi-sheet Excel file with realistic actuarial data:

   • Life tables with mortality rates, survival probabilities, and life expectancy (101 ages)
   • Insurance policies with demographics, premiums, face amounts (1,000 policies)
   • Claims data with types, amounts, status, and investigation times (150 claims)
   • Reserve calculations by product type and valuation year (25 records)

2. Converts each sheet to multiple formats:

   • Parquet files for efficient data analysis
   • CSV files for human readability and Excel compatibility
   • JSON files for web applications and APIs
   • Schema files with metadata, statistics, and data types

3. Generates comprehensive analytics:

   • Summary statistics and data quality reports
   • Loss ratio analysis by policy type
   • Mortality analysis and high-risk identification
   • Reserve adequacy and trend analysis

Advanced Analytics Example

Perform comprehensive actuarial analysis on the converted data:

python3 analyze_polars_data.py

Features include:

• Claims Analysis: By status, type, and high-value claims
• Policy Analysis: Premium rates, face amounts, and status distribution
• Mortality Analysis: High mortality ages and life expectancy statistics
• Reserve Analysis: Trends by year and product type
• Loss Ratios: Policy-to-claim analysis with advanced joins
• Data Export: Filtered datasets and summary reports

Project Structure

├── excel_polars_mcp/          # Core MCP server implementation
│   ├── __init__.py
│   └── server.py              # FastMCP server with Excel conversion tools
├── examples/                  # Example scripts and demos
│   ├── demo.py               # Basic usage demonstration
│   ├── create_actuarial_data.py  # Generate sample actuarial Excel file
│   └── convert_actuarial_data.py # Convert Excel to Polars formats
├── sample_data/               # Generated sample data
│   └── actuarial_data.xlsx   # Multi-sheet actuarial Excel file
├── output/                    # Converted data in multiple formats
│   ├── *.parquet             # Efficient binary format
│   ├── *.csv                 # Human-readable format
│   ├── *.json                # Structured data format
│   ├── *_schema.json         # Metadata and statistics
│   └── conversion_summary.md  # Detailed conversion report
├── tests/                     # Comprehensive test suite
├── run_actuarial_example.py  # One-command actuarial demo
├── analyze_polars_data.py    # Advanced analytics demonstration
├── view_actuarial_data.py    # Data inspection utility
└── pyproject.toml            # Project configuration with uv

## Development

Install development dependencies:

```bash
uv sync --dev

Run tests:

uv run pytest

Format code:

uv run black .
uv run ruff check --fix .

Type checking:

uv run mypy .

Quick Start

1. Clone and setup:

   git clone <repository>
   cd python-mcp-excel-polars
   uv sync
   

2. Run the actuarial example:

   python3 run_actuarial_example.py
   

3. Analyze the results:

   python3 analyze_polars_data.py
   

4. Explore the data:

   python3 view_actuarial_data.py
   

Dependencies

• Polars: High-performance DataFrames
• FastMCP: Model Context Protocol implementation
• openpyxl: Excel file reading/writing
• xlsxwriter: Multi-sheet Excel creation
• fastexcel: Optimized Excel processing

Use Cases

• Actuarial Analysis: Life tables, mortality studies, reserve calculations
• Insurance Analytics: Policy analysis, claims processing, loss ratios
• Data Migration: Excel to modern data formats (Parquet, JSON)
• Financial Modeling: Risk assessment, statistical analysis
• Business Intelligence: Automated reporting and data transformation

License

MIT License - see LICENSE file for details.