load_datasetLABSCO SUMMARY
The quick-data analytics set, with the Windows problems sorted out.
This is disler/quick-data-mcp with Windows file paths, Claude Desktop launchers and dependency installation fixed, by an author who says openly that it is his first project of this kind. If you are on Windows and the original would not start, that is exactly the gap it closes.
What it is
A Windows-focused fork of quick-data-mcp for Claude Desktop: analytics over any JSON or CSV file, with Windows path handling and batch launchers.
What you get
- A dataset loaded from any CSV or JSON file, whatever its structure (load_dataset)
- Automatic schema discovery, with analyses suggested from what it finds
- The datasets currently in memory, and details for one of them (list_loaded_datasets, get_dataset_info)
- A loaded dataset broken down by a column (segment_by_column)
- Memory released when you are finished with a dataset (clear_dataset)
- Interactive Plotly charts, and an analytics set that runs from basic statistics upward
Requirements
Windows 10 or newer, Python 3.9 or newer and Claude Desktop. Cloned from the repository, then install_dependencies.bat.
Setup effort
Build from source — git clone https://github.com/Beaulewis1977/quick-data-for-windows-mcp.git
USE CASES
01
Analysing a spreadsheet without opening a notebook
Load the file and let schema discovery tell you what is in it and what to ask.
02
Finding out what is in a file someone sent you
Load it and read back the columns, types and suggested analyses.
03
Breaking results down by category
Segment the loaded dataset by the column that matters and compare the groups.
04
Working across several files in one session
Load each one and check what is currently in memory before comparing them.
05
Freeing memory after a large file
Clear the dataset when you are done rather than restarting the client.
17 tools
Listed in project docs. Your client may show a different list.
list_loaded_datasetsget_dataset_infoclear_datasetclear_all_datasetssegment_by_columnfind_correlationsanalyze_distributionsdetect_outlierssuggest_analysiscreate_chartgenerate_dashboardvalidate_data_qualitycompare_datasetsmerge_datasetscalculate_feature_importanceexport_insightsBEFORE YOU INSTALL
What to know first
Tools you get
17 tools named in the project's documentation.
Requires
Python 3.9+
Where it runs
Both documented setups run the server on your machine.
INSTALLATION
Connect your client
You clone this one yourself.
Server setup
TERMINAL
git clone https://github.com/Beaulewis1977/quick-data-for-windows-mcp.git
cd quick-data-for-windows-mcp
install_dependencies.batMCP Clients
Claude Desktop — Python
CLAUDE DESKTOP — PYTHON · JSON
{
"mcpServers": {
"quick-data-windows": {
"command": "python",
"args": [
"<YOUR_PATH_TO_QUICK_DATA_FOR_WINDOWS>\\main.py"
],
"cwd": "<YOUR_PATH_TO_QUICK_DATA_FOR_WINDOWS>",
"env": {
"LOG_LEVEL": "INFO",
"PYTHONPATH": "<YOUR_PATH_TO_QUICK_DATA_FOR_WINDOWS>\\src"
}
}
}
}Claude Desktop — uv
CLAUDE DESKTOP — UV · JSON
{
"mcpServers": {
"quick-data-windows": {
"command": "uv",
"args": [
"--directory",
"<YOUR_PATH_TO_QUICK_DATA_FOR_WINDOWS>",
"run",
"python",
"main.py"
]
}
}
}
