Hugging Face MCP Server

Example Use Cases

• 1

  Query and explore machine learning datasets from the Hugging Face Hub.

• 2

  Perform statistical analysis and data filtering on datasets.

• 3

  Detect PII entities and validate dataset integrity.

Description

This MCP server offers 12 tools that cover all GET endpoints of the Hugging Face Dataset Viewer API, enabling users to query datasets, explore data, perform full-text searches, analyze statistics, and validate datasets from the Hugging Face Hub. It supports features like SQL-like filtering, PII entity detection, and parquet file information retrieval. The server is built with FastMCP and can be run locally or via Docker, facilitating easy integration with Claude Desktop and other environments.