Vibe Preprocessing and Analysis MCP Server for CSV files
About
A powerful MCP (Model Control Protocol) server for preprocessing and analyzing CSV files. This server provides a suite of tools for data manipulation, visualization, and analysis.
Details
- Author
- mudit14224
- Downloads
- 308
- Categories
- Search
Jump to
- Load CSV files and manage working directories
- Handle null values with multiple strategies (remove, fill, forward/backward fill)
- Drop and rename columns, run custom DataFrame editing code
- Generate statistical summaries and correlation matrices with visualizations
- Create nine types of plots (line, bar, scatter, histogram, box, violin, pie, count, KDE)
- Save processed DataFrames and visualizations to the working directory
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Vibe Preprocessing and Analysis MCP Server for CSV filesCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Install dependencies with uv add "mcp[cli]" pandas matplotlib seaborn numpy (or pip), then run mcp install server.py to register the server in Claude Desktop. Use mcp dev server.py to test with the MCP Inspector. Set the working directory via the set_work_dir tool or the WORK_DIR environment variable. Use the provided tools in natural language or through the inspector to load, preprocess, analyze, and visualize CSV files.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"vibe preprocessing and analysis mcp server for csv files": {
"Vibe-Data-Analysis": {
"command": "uv",
"args": [
"run",
"mcp"
]
}
}
}
}McpServers
{
"Vibe-Data-Analysis": {
"command": "uv",
"args": [
"run",
"mcp"
]
}
}Vibe Preprocessing and Analysis MCP Server for CSV files
A powerful MCP (Model Control Protocol) server for preprocessing and analyzing CSV files. This server provides a suite of tools for data manipulation, visualization, and analysis through an intuitive interface.
Features
- Data Loading and Management
- Load CSV files from a specified working directory
- Set and manage working directories
- List files in the working directory
- Save processed dataframes to new files
- Data Preprocessing
- Handle mixed data types in columns
- Manage null values with various strategies:
- Remove rows with nulls
- Fill with mean/median/mode
- Forward/backward fill
- Fill with constant values
- Drop and rename columns
- Run custom dataframe editing code
- Save processed data to new files
- Data Analysis
- Generate comprehensive data descriptions
- Create correlation matrices with visualizations
- Handle mixed data types in columns
- Run custom analysis code
- Data Visualization
- Create various types of plots:
- Line plots
- Bar charts
- Scatter plots
- Histograms with KDE
- Box plots
- Violin plots
- Pie charts
- Count plots
- Kernel Density Estimation plots
- Custom graph generation through code
- Save visualizations to the working directory
- Run custom visualization code
Setup Instructions
Prerequisites
- Python 3.x - uv (recommended package manager). I recommend using uv to manage the server.Installation
1. Add MCP and required dependencies:uv add "mcp[cli]"
uv add pandas matplotlib seaborn numpy
2. Install the server in Claude Desktop:
mcp install server.py
Alternative Installation with pip
If you prefer using pip:pip install "mcp[cli]" pandas matplotlib seaborn numpy
Usage
1. Start the MCP server:
uv run mcp
2. Test the server using MCP Inspector:
mcp dev server.py
You can install this server in Claude Desktop and interact with it right away by running:
mcp install server.py
Alternatively, you can test it with the MCP Inspector:
mcp dev server.py
Available Tools
Data Management
-send_work_dir(): Retrieve the current working directory
- set_work_dir(new_work_dir): Set a new working directory
- list_work_dir_files(): List files in the current working directory
- load_csv(filename): Load a CSV file into the system
- save_global_df(filename): Save the current dataframe to a file
Data Preprocessing
-handle_column_mixed_types(): Handle columns with mixed data types
- handle_null_values(strategy, columns): Handle null values in the dataset with various strategies
- drop_columns(columns): Remove specified columns
- rename_columns(column_mapping): Rename columns in the dataframe
- run_custom_df_edit_code(code): Execute custom dataframe manipulation code
Data Analysis
-describe_df(): Generate a statistical summary of the dataframe
- generate_correlation_matrix(): Create a correlation matrix with visualization
Data Visualization
-plot_graph(graph_type, x_column, y_column, output_filename): Create various types of plots
- Supported graph types: line, bar, scatter, hist, box, violin, pie, count, kde
- run_custom_graph_code(code): Execute custom visualization code
Environment Variables
- WORK_DIR: The working directory where files are read from and saved to
Error Handling
The server includes comprehensive error handling for:
- Missing working directories
- File not found errors
- Data loading and processing errors
- Invalid operations on empty dataframes
- Mixed data type handling
- Custom code execution errors
- Invalid column names
- Invalid graph types
- Null value handling errors
Contributing
Feel free to submit issues and enhancement requests!
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
