Dataset Viewer

🚀 Dataset Viewer MCP Server

A MCP server for interacting with the Hugging Face Dataset Viewer API, which provides the functionality to browse and analyze datasets hosted on Hugging Face repositories.

✨ Features

Resources

• Access Hugging Face datasets using the dataset:// URI scheme.
• Support dataset configurations and splits.
• Provide paginated access to dataset content.
• Handle authentication for private datasets.
• Support searching and filtering of dataset content.
• Provide dataset statistics and analysis.

Tools

The server offers the following tools:

1. validate

   • Check if a dataset exists and is accessible.
   • Parameters:
     • dataset: Dataset identifier (e.g., 'stanfordnlp/imdb').
     • auth_token (optional): For private datasets.

2. get_info

   • Get detailed information about a dataset.
   • Parameters:
     • dataset: Dataset identifier.
     • auth_token (optional): For private datasets.

3. get_rows

   • Get paginated access to dataset content.
   • Parameters:
     • dataset: Dataset identifier.
     • config: Configuration name.
     • split: Split name.
     • page (optional): Page number (starting from 0).
     • auth_token (optional): For private datasets.

4. get_first_rows

   • Get the first few rows of a dataset split.
   • Parameters:
     • dataset: Dataset identifier.
     • config: Configuration name.
     • split: Split name.
     • auth_token (optional): For private datasets.

5. get_statistics

   • Get statistics for a dataset split.
   • Parameters:
     • dataset: Dataset identifier.
     • config: Configuration name.
     • split: Split name.

6. search

   • Search for content in a dataset.
   • Parameters:
     • dataset: Dataset identifier.
     • query: Search query.
     • auth_token (optional): For private datasets.

7. filter

   • Filter dataset content based on conditions.
   • Parameters:
     • dataset: Dataset identifier.
     • condition: Filter condition.
     • auth_token (optional): For private datasets.

📦 Installation

Clone the repository using Git

git clone https://github.com/your-organization/dataset-viewer-mcp.git
cd dataset-viewer-mcp

Install dependencies

npm install

Start the server

node index.js

Configure environment variables

Set the Hugging Face API token

Set the following environment variable to access Hugging Face's API:

export HUGGING_FACE_TOKEN=your_api_token_here

Configuration on Windows

On Windows systems, you can use the following command to set the environment variable:

set HUGGING_FACE_TOKEN=your_api_token_here

Configuration example - Claude Desktop

To configure on Claude Desktop, add the following content to the corresponding configuration file:

{
  "datasetViewerMcp": {
    "path": "%应用程序数据%/claude-desktop/dataset-viewer-mcp",
    "configPath": "~/.config/claude/dataset-viewer-mcp.json"
  }
}

💻 Usage Examples

Example 1: Validate dataset existence

{
  "dataset": "stanfordnlp/imdb",
  "auth_token": "your_api_token_here"
}

Example 2: Get dataset information

{
  "dataset": "stanfordnlp/imdb",
  "auth_token": "your_api_token_here"
}

Example 3: Get paginated content

{
  "dataset": "stanfordnlp/imdb",
  "config": "default",
  "split": "train",
  "page": 0,
  "auth_token": "your_api_token_here"
}

Example 4: Search dataset content

{
  "dataset": "stanfordnlp/imdb",
  "query": "电影评论",
  "auth_token": "your_api_token_here"
}

📄 License

MIT License - See LICENSE for details.