AI server for CSV processing with 40+ specialized tools for data manipulation, analysis, and validation featuring autosave, undo/redo capabilities, and support for gigabyte-sized files.

Installation

Smithery (Recommended)

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

UV (Fastest)

curl -LsSf https://astral.sh/uv/install.sh | sh
git clone https://github.com/santoshray02/csv-editor.git
cd csv-editor
uv sync
uv run csv-editor

From Source

git clone https://github.com/santoshray02/csv-editor.git
cd csv-editor
pip install -e .

Pipx (Global)

pipx install git+https://github.com/santoshray02/csv-editor.git

GitHub

pip install git+https://github.com/santoshray02/csv-editor.git

Configuration

Claude Desktop

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

Available Tools

Tool	Description
load_csv	Load from file
load_csv_from_url	Load from URL
load_csv_from_content	Load from string
export_csv	Export to various formats
get_session_info	Session details
list_sessions	Active sessions
close_session	Cleanup
filter_rows	Advanced filtering
sort_data	Sort by multiple columns
select_columns	Column selection
rename_columns	Rename columns
add_column	Add computed columns
remove_columns	Remove columns
update_column	Update values
change_column_type	Type conversion

Features

• 40+ specialized tools for CSV processing
• Load and export in multiple formats (CSV, JSON, Excel, Parquet, HTML, Markdown)
• Data transformation (filtering, sorting, grouping, pivot tables, joining)
• Data cleaning (deduplication, missing value handling, type fixing)
• Statistical analysis and correlation matrices
• Outlier detection using IQR and Z-score methods
• Data quality assessment and validation
• Automatic saving with customizable strategies
• Full undo/redo with operation history tracking
• Multi-user session management with isolation

Environment Variables

Optional

• CSV_MAX_FILE_SIZE - Maximum file size (default: 1GB)
• CSV_SESSION_TIMEOUT - Session timeout in seconds (default: 3600s)
• CSV_CHUNK_SIZE - Chunk size for processing (default: 10000)
• CSV_AUTO_SAVE - Enable autosave (default: true)

Usage Examples

Load sales data and remove duplicates

Filter Q4 2024 transactions over $10,000

Calculate correlation between price and quantity

Fill missing values with median

Export as Excel with analysis

Notes

Built on FastMCP and Pandas. Supports autosave strategies (overwrite, backup, versioning, custom). Includes comprehensive data quality assurance tools. See MCP_CONFIG.md for configuration with other clients (Continue, Cline, Windsurf, Zed).