Back to the MCP Gateway catalog

MCP Gateway · Catalog

Hugging Face icon

Hugging Face MCP server

Official

Hugging Face MCP is the official Model Context Protocol server for Hugging Face, with OAuth and tools that let agents search models, Spaces, papers, and docs, and run Gradio applications. Connect it through the Speakeasy MCP Gateway and agents such as Claude, Cursor, and ChatGPT get one governed URL for it, with SSO, RBAC, and a full audit trail.

AI & search10 toolsAuth: oauth

Behind the MCP Gateway

Run Hugging Face MCP behind the MCP Gateway

One URL for every agent, sign-in through your IdP, RBAC down to the tool, and every Hugging Face call inspected and logged.

Deploy on MCP Gateway

Tools · 10

Each tool below can be allowed or denied per team and role at the MCP Gateway, and every call is inspected and logged.

hf_whoami

Hugging Face tools are being used by authenticated user 'proctorpmcp'

space_search

Find Hugging Face Spaces using semantic search. IMPORTANT Only MCP Servers can be used with the dynamic_space toolInclude links to the Space when presenting the results.

hub_repo_search

Search Hugging Face repositories with a shared query interface. You can target models, datasets, spaces, or aggregate across multiple repo types in one call. Use space_search for semantic-first discovery of Spaces. Include links to repositories in your response.

paper_search

Find Machine Learning research papers on the Hugging Face hub. Include 'Link to paper' When presenting the results. Consider whether tabulating results matches user intent.

hub_repo_details

Get details for one or more Hugging Face repos (model, dataset, or space). Auto-detects type unless specified.

hf_doc_search

Search and Discover Hugging Face Product and Library documentation. Send an empty query to discover structure and navigation instructions. Knowledge up-to-date as at 8 April 2026. Combine with the Product filter to focus results.

hf_doc_fetch

Fetch a document from the Hugging Face or Gradio documentation library. For large documents, use offset to get subsequent chunks.

dynamic_space

Perform Tasks with Hugging Face Spaces. Use "discover" to view available Tasks. Examples are Image Generation/Editing, Background Removal, Text to Speech, OCR and many more. Call with no arguments for full usage instructions.

hf_hub_query

Read-only Hugging Face Hub navigator for discovery, lookup, filtering, ranking, counts, field-constrained extraction, and relationship questions across users, orgs, models, datasets, spaces, collections, discussions, daily papers, recent activity, followers/following, likes, and likers. Good for structured raw outputs and compact results. Generated helper calls can explicitly bound limit, scan_limit, max_pages, and ranking_window for brevity or broader coverage, and the tool can also be asked about its supported helpers, canonical fields, defaults, and coverage behavior.

gr1_z_image_turbo_generate

Generate an image using the Z-Image model based on the provided prompt and settings. This function is triggered when the user clicks the "Generate" button. It processes the input prompt (optionally enhancing it), configures generation parameters, and produces an image using the Z-Image diffusion transformer pipeline. Returns: tuple: (gallery_images, seed_str, seed_int), - seed_str: String representation of the seed used for generation, - seed_int: Integer representation of the seed used for generation (from mcp-tools/Z-Image-Turbo)

Questions

What is a Hugging Face MCP server?
A Hugging Face MCP server exposes Hugging Face to AI agents as Model Context Protocol tools. Instead of a custom integration per agent, any MCP client can connect to the server and call tools such as hf_whoami or space_search to work with Hugging Face. The server in this catalog is the official Hugging Face MCP server, which authenticates with OAuth.
What can you do with Hugging Face MCP?
Agents can search models, Spaces, papers, and docs, and run Gradio applications. The catalog lists 10 tools for this server, including hf_whoami, space_search, and hub_repo_search. The full list with descriptions is on this page.
Why put Hugging Face MCP behind an MCP gateway?
Connecting the Hugging Face MCP server directly to each client means a separate credential per developer, no inventory of who can reach Hugging Face through an agent, and no record of what agents did with it. An MCP gateway sits between every agent and the Hugging Face server, so sign-in runs through your identity provider, role-based access decides which teams can call which Hugging Face tools, and every call is inspected and logged in one place. The Speakeasy MCP Gateway applies those controls to Hugging Face and to every other server in your catalog.
How do you connect Hugging Face MCP through a gateway?
Add the Hugging Face server from the Speakeasy MCP catalog and every agent reaches it through one gateway URL, next to the rest of your approved servers. Authentication uses OAuth 2.1 with PKCE and dynamic client registration, sign-in runs through your identity provider, and role-based access controls decide which teams can use which Hugging Face tools. Every prompt, response, and tool call is inspected and logged at the gateway.
How is this different from an unofficial Hugging Face MCP?
Community-built Hugging Face MCP servers wrap the same Hugging Face surface but are not vendor-maintained, and most authenticate with a static credential that gives an agent everything the credential owner can touch. This catalog lists the official Hugging Face MCP server, which uses OAuth. Running it behind the Speakeasy MCP Gateway adds SSO, role-based access, and an audit trail of every call.

AI everywhere.

Control here.

Talk to usLearn more