pandas-mcp-server
If you are the rightful owner of pandas-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcpreview.com.
Pandas MCP Server is a server implementation using the Model Context Protocol (MCP) to handle CSV files, execute Pandas code, and generate interactive charts.
The Pandas MCP Server is designed to facilitate data manipulation and visualization through a server that leverages the Model Context Protocol (MCP). It provides functionalities to read metadata from CSV files, execute Pandas code securely, and generate interactive charts such as bar, pie, and line charts using Chart.js. The server is built to handle large datasets efficiently, with a file size limit of 100MB to prevent excessive memory usage. It also includes security measures to prevent the execution of potentially harmful operations. The server is compatible with Python 3.11 or higher and requires the installation of necessary packages via a requirements file. Users can interact with the server through an MCP client, allowing for seamless integration into data analysis workflows.
Features
- CSV Metadata Reading: Loads a CSV file and returns its column structure and sample data, with automatic detection of file encoding and delimiter.
- Secure Pandas Code Execution: Executes Pandas code provided as a string, with security measures to prevent execution of blacklisted operations.
- Interactive Bar Chart Generation: Generates an interactive HTML bar chart using Chart.js, with customizable categories, values, and title.
- Interactive Pie Chart Generation: Creates an interactive HTML pie chart using Chart.js, with customizable labels, values, and title.
- Interactive Line Chart Generation: Produces an interactive HTML line chart using Chart.js, with customizable labels, datasets, and title.