csv-editor

Overview

CSV Editor is a comprehensive MCP server that provides AI assistants with full CSV editing capabilities through 39 specialized tools. Built on FastMCP 3 and powered by pandas, it enables stateful CSV operations with advanced features like sessions, undo/redo functionality, and configurable auto-save strategies. Unlike read-only data MCPs, this server focuses on allowing AI to make actual changes to CSV data with proper tracking and history management. It supports various input/output formats including CSV, JSON, Excel, Parquet, HTML, and Markdown files.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• load_csv

  Load CSV data from a file

  Try:Load sales.csvTry:Load customer_data.csv
• filter_rows

  Filter rows based on conditions

  Try:Filter rows where status is activeTry:Show transactions greater than $1000
• get_statistics

  Get statistical summary of data

  Try:Calculate statistics for revenue columnTry:Show descriptive statistics for the dataset
• add_column

  Add computed columns to the dataset

  Try:Add a profit column as revenue minus costTry:Create a full_name column combining first and last name
• export_csv

  Export data to various formats

  Try:Export to Excel formatTry:Save the data as a JSON file
• remove_duplicates

  Remove duplicate rows from the dataset

  Try:Remove duplicate entries based on emailTry:Clean the data by removing duplicates
• fill_missing_values

  Handle missing values in the dataset

  Try:Fill missing values with the meanTry:Replace nulls in the age column with zero
• get_correlation_matrix

  Calculate correlation between columns

  Try:Find correlation between price and quantityTry:Show correlation matrix for numeric columns
• group_by_aggregate

  Group data and perform aggregations

  Try:Group by region and sum salesTry:Calculate average price per category
• detect_outliers

  Detect outliers in the data

  Try:Find outliers in the sales columnTry:Detect anomalies using IQR method
• undo

  Undo the last operation

  Try:Undo the last changeTry:Revert the previous operation
• check_data_quality

  Check data quality and generate a report

  Try:Analyze data quality issuesTry:Check for missing values and duplicates

Comparable tools

Installation

npx -y @smithery/cli install @santoshray02/csv-editor --client claude

# Install uv if needed (one-time setup)
curl -LsSf https://astral.sh/uv/install.sh | sh

# Clone and run
git clone https://github.com/santoshray02/csv-editor.git
cd csv-editor
uv sync
uv run csv-editor

{
  "mcpServers": {
    "csv-editor": {
      "command": "uv",
      "args": ["tool", "run", "csv-editor"],
      "env": {
        "CSV_MAX_FILE_SIZE": "1073741824"
      }
    }
  }
}

FAQ

What makes CSV Editor different from other CSV/data MCP servers?

CSV Editor focuses on stateful editing with session management, undo/redo functionality, and auto-save capabilities - most other CSV MCPs are read-only or single-shot analysis tools.

What file formats does CSV Editor support?

It supports CSV, JSON, Excel, Parquet, HTML, and Markdown for both input and output operations, making it versatile for different data processing needs.

Compare csv-editor with