Zum Inhalt springen
Kitbay

hf-mcp

Apache-2.0Aktualisiert am 11. Nov. 202611.334GitHub-Sterne

A Hugging Face skill that teaches an agent how to use the tools of the Hugging Face MCP server once a client is connected to it.

Was der Skill vermittelt

A Hugging Face skill that teaches an agent how to use the tools of the Hugging Face MCP server once a client is connected to it. Worked examples show which tool calls to chain for common requests: searching models, datasets, Spaces and papers, comparing models from different authors, reading repository details and READMEs, fetching Hugging Face documentation, running compute jobs and calling Gradio Spaces as tools. A tool selection guide and tips round it out. The skill does nothing without the MCP connection, which is set up from the Hugging Face MCP settings page.

Funktionen

  • Search Hub models, datasets, Spaces and papers via MCP
  • Read repository details and model cards
  • Fetch Hugging Face documentation
  • Run compute jobs and Gradio Spaces as tools