`search_datasets`

Pack: huggingface · Connect: https://pipeworx.io/mcp (see Connect below for a single-pack URL)

No MCP client? Call it directly: GET https://gateway.pipeworx.io/v1/tools/search_datasets for the schema, then POST the same URL with its arguments for the data.

Search Hugging Face Hub datasets with filters for author, language, and task_categories; sort by downloads, likes, or lastModified; returns dataset id, downloads, likes, and tags.

Parameters

Name	Type	Required	Description
`search`	string	no
`author`	string	no
`language`	string	no	ISO language code, e.g. “fr”
`task_categories`	string	no	Comma-separated task categories, e.g. “question-answering”
`sort`	string	no
`direction`	string	no
`limit`	number	no
`full`	boolean	no

Example call

Arguments

{
  "search": "imagenet",
  "limit": 20
}

curl

curl -X POST https://gateway.pipeworx.io/huggingface/mcp \
  -H 'Content-Type: application/json' \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"search_datasets","arguments":{"search":"imagenet","limit":20}}}'

TypeScript (@pipeworx/sdk)

import { Pipeworx } from '@pipeworx/sdk';
const pipeworx = new Pipeworx();

const result = await pipeworx.call('search_datasets', {
  "search": "imagenet",
  "limit": 20
});

More examples

{
  "author": "wikipedia",
  "language": "en",
  "task_categories": "text-classification",
  "sort": "likes",
  "limit": 15
}

Response shape

Full JSON Schema

{
  "type": "object",
  "description": "List of datasets matching search criteria"
}

Connect

Add this to your MCP client config — every tool in the catalog, including this one — or use one-click install buttons:

{
  "mcpServers": {
    "pipeworx": {
      "url": "https://pipeworx.io/mcp"
    }
  }
}

Connect to just the huggingface pack

{
  "mcpServers": {
    "huggingface": {
      "url": "https://gateway.pipeworx.io/huggingface/mcp"
    }
  }
}

See Getting Started for client-specific install steps.

search_datasets