Check if a dataset exists and is accessible
Get detailed information about a dataset
Get paginated contents of a dataset
Get first rows from a dataset split
Get statistics about a dataset split
Search for text within a dataset
Filter rows using SQL-like conditions
Download entire dataset in Parquet format
The Hugging Face Dataset Viewer MCP provides a seamless interface to explore, search, and analyze datasets hosted on the Hugging Face Hub. It enables users to validate datasets, retrieve detailed information, access paginated contents, and perform advanced operations like searching and filtering. With support for dataset configurations, splits, and authentication for private datasets, this MCP offers comprehensive capabilities for data exploration. It also provides statistical analysis and the ability to download entire datasets in Parquet format, making it an essential tool for data scientists and machine learning practitioners.
git clone https://github.com/privetin/dataset-viewer.git
cd dataset-viewer
# Create virtual environment
uv venv
# Activate virtual environment
# On Unix:
source .venv/bin/activate
# On Windows:
.venv\Scripts\activate
uv add -e .
You can set the HUGGINGFACE_TOKEN environment variable to provide your Hugging Face API token for accessing private datasets.
Add the MCP server configuration to your Claude Desktop config file:
%APPDATA%\Claude\claude_desktop_config.json~/Library/Application Support/Claude/claude_desktop_config.json{
"mcpServers": {
"dataset-viewer": {
"command": "uv",
"args": [
"--directory",
"PATH_TO_YOUR_DATASET_VIEWER_DIRECTORY",
"run",
"dataset-viewer"
]
}
}
}
Replace PATH_TO_YOUR_DATASET_VIEWER_DIRECTORY with the actual path to where you cloned the repository.
Once installed, you can use the Dataset Viewer MCP to interact with Hugging Face datasets. The MCP uses the dataset:// URI scheme for accessing datasets.
For private datasets, you'll need to provide an authentication token either through the environment variable or as a parameter to the relevant tools.