Summary
Connects Claude to the Hugging Face Hub API for browsing and inspecting models, datasets, and Spaces. You get search across all three categories, detailed metadata retrieval, file listings at the repo root, and trending feeds. Useful when you need Claude to recommend models for a task, pull dataset documentation, or check what files ship with a particular checkpoint. Part of the Pipeworx gateway, so you can either connect to this standalone endpoint or use the full gateway with its ask_pipeworx natural language wrapper that routes your questions to the right tool automatically.
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
Tools
Public tool metadata for what this MCP can expose to an agent.
8 tools
hf_whoamiHugging Face tools are being used anonymously and may be rate limited. Call this tool for instructions on joining and authenticating.
Hugging Face tools are being used anonymously and may be rate limited. Call this tool for instructions on joining and authenticating.
No parameter schema in public metadata yet.
space_searchFind 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.3 params
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.
Parameters* required
limitnumberNumber of results to returndefault: 10
mcpbooleanOnly return MCP Server enabled Spacesdefault: false
querystringSemantic Search Query
hub_repo_searchSearch 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.6 params
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.
Parameters* required
authorstringOrganization or user namespace to filter by (e.g. 'google', 'meta-llama', 'huggingface').
filtersarrayOptional hub filter tags. Applied to each selected repo type (e.g. ["text-generation"], ["language:en"], ["mcp-server"]).
limitnumberMaximum number of results to return per selected repo typedefault: 20
querystringSearch term. Leave blank and specify sort + limit to browse trending or recent repositories.
repo_typesarrayRepository types to search. Defaults to ["model", "dataset"]. space uses keyword search via /api/spaces.
sortstringSort order (descending): trendingScore, downloads, likes, createdAt, lastModifiedone of
trendingScore · downloads · likes · createdAt · lastModifiedpaper_searchFind Machine Learning research papers on the Hugging Face hub. Include 'Link to paper' When presenting the results. Consider whether tabulating results matches user intent.3 params
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.
Parameters* required
concise_onlybooleanReturn a 2 sentence summary of the abstract. Use for broad search terms which may return a lot of results. Check with User if unsure.default: false
querystringSemantic Search query
results_limitnumberNumber of results to returndefault: 12
hub_repo_detailsGet details for one or more Hugging Face repos (model, dataset, or space). Auto-detects type unless specified.2 params
Get details for one or more Hugging Face repos (model, dataset, or space). Auto-detects type unless specified.
Parameters* required
repo_idsarrayRepo IDs for (models|dataset/space) - usually in author/name format (e.g. openai/gpt-oss-120b)
repo_typestringSpecify lookup type; otherwise auto-detectsone of
model · dataset · spacehf_doc_searchSearch and Discover Hugging Face Product and Library documentation. Send an empty query to discover structure and navigation instructions. Knowledge up-to-date as at 11 March 2026. Combine with the Product filter to focus results.2 params
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 11 March 2026. Combine with the Product filter to focus results.
Parameters* required
productstringFilter by Product. Supply when known for focused results
querystringStart with an empty query for structure, endpoint discovery and navigation tips. Use semantic queries for targetted searches.
hf_doc_fetchFetch a document from the Hugging Face or Gradio documentation library. For large documents, use offset to get subsequent chunks.2 params
Fetch a document from the Hugging Face or Gradio documentation library. For large documents, use offset to get subsequent chunks.
Parameters* required
doc_urlstringDocumentation URL (Hugging Face or Gradio)
offsetnumberToken offset for large documents (use the offset from truncation message)
gr1_z_image_turbo_generateGenerate 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 th...6 params
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 th...
Parameters* required
promptstringText prompt describing the desired image content
random_seedbooleanWhether to generate a new random seed, if True will ignore the seed inputdefault: true
resolutionstringOutput resolution in format "WIDTHxHEIGHT ( RATIO )" (e.g., "1024x1024 ( 1:1 )")one of
1024x1024 ( 1:1 ) · 1152x896 ( 9:7 ) · 896x1152 ( 7:9 ) · 1152x864 ( 4:3 ) · 864x1152 ( 3:4 ) · 1248x832 ( 3:2 )default: 1024x1024 ( 1:1 )seedintegerSeed for reproducible generationdefault: 42
shiftnumberTime shift parameter for the flow matching schedulerdefault: 3
stepsnumberNumber of inference steps for the diffusion processdefault: 8
@pipeworx/huggingface
Hugging Face Hub MCP — search and inspect models, datasets, and Spaces on the largest open ML repository. No auth required for public reads.
Part of Pipeworx — an MCP gateway connecting AI agents to 1476+ live data sources.
Tools
search_models(search?, author?, library?, language?, pipeline_tag?, tags?, sort?, direction?, limit?, full?)search_datasets(search?, author?, language?, task_categories?, sort?, direction?, limit?, full?)search_spaces(search?, author?, sdk?, sort?, direction?, limit?, full?)get_model(repo_id, revision?)get_dataset(repo_id, revision?)get_space(repo_id, revision?)list_model_files(repo_id, revision?)/list_dataset_files(repo_id, revision?)trending_models(limit?)/trending_datasets(limit?)
Auth
Public reads are keyless. Higher rate limits and gated/private repos require a token — optional:
- Platform key (optional): gateway env
PLATFORM_HF_KEY - BYO (optional):
?_apiKey=<hf_token>
Data source
https://huggingface.co/api/ — public REST.
Quick Start
Add to your MCP client (Claude Desktop, Cursor, Windsurf, etc.):
{
"mcpServers": {
"huggingface": {
"url": "https://gateway.pipeworx.io/huggingface/mcp"
}
}
}
What this endpoint actually serves
tools/list at https://gateway.pipeworx.io/huggingface/mcp returns the tools in the table
above plus the shared Pipeworx meta-tools — ask_pipeworx,
discover_tools, search_within, remember/recall and the rest of the
gateway-wide set. So the tool count you see is larger than this table: a
single-pack endpoint currently lists roughly 30 shared tools alongside the
pack's own. The connection's initialize response states its exact scope, and
is the authoritative answer for a given day.
This is deliberate, not multiplexing by accident. The meta-tools are what let a
scoped connection answer a question this pack does not cover — via
ask_pipeworx, which routes across the whole catalog — without you adding a
second MCP server. There is currently no way to mount a pack endpoint without
them; if the extra schemas cost you more context than the routing is worth,
connect to the full gateway once rather than to several pack endpoints.
Or connect to the full Pipeworx gateway to get every pack's tools listed directly, instead of just this one's:
{
"mcpServers": {
"pipeworx": {
"url": "https://gateway.pipeworx.io/mcp"
}
}
}
Both URLs reach the same gateway and the same 1476+ data sources. The
only difference is which pack's tools are listed directly; ask_pipeworx
reaches all of them from either one.
Using with ask_pipeworx
Instead of calling tools directly, you can ask questions in plain English — this works on the pack endpoint above as well as on the full gateway:
ask_pipeworx({ question: "your question about Huggingface data" })
The gateway picks the right tool and fills the arguments automatically.
More
License
MIT
Featured
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
