hf-mcp

$npx mdskill add huggingface/skills/hf-mcp

Search, inspect, and utilize models, datasets, and tools directly from the Hugging Face Hub.

  • Find suitable models, datasets, and interactive AI applications for specific tasks.
  • Integrates with the Hugging Face Hub via a dedicated MCP server connection.
  • Executes searches, retrieves repository metadata, and runs compute jobs on demand.
  • Delivers structured results including documentation, model details, and functional Spaces.

SKILL.md

.github/skills/hf-mcpView on GitHub ↗
---
name: hf-mcp
description: Use Hugging Face Hub via MCP server tools. Search models, datasets, Spaces, papers. Get repo details, fetch documentation, run compute jobs, and use Gradio Spaces as AI tools. Available when connected to the HF MCP server.
---

# Hugging Face MCP Server

Connect AI assistants to the Hugging Face Hub. Setup: https://huggingface.co/settings/mcp

## Use Cases & Examples

### Find the Best Model for a Task

```
User: "Find the best model for code generation"

1. model_search(task="text-generation", query="code", sort="trendingScore", limit=10)
2. hub_repo_details(repo_ids=["top-result-id"], include_readme=true)
```

### Compare Models from Different Providers

```
User: "Compare Llama vs Qwen for text generation"

1. model_search(author="meta-llama", task="text-generation", sort="downloads", limit=5)
2. model_search(author="Qwen", task="text-generation", sort="downloads", limit=5)
3. hub_repo_details(repo_ids=["meta-llama/Llama-3.2-1B", "Qwen/Qwen3-8B"], include_readme=true)
```

### Find Training Datasets

```
User: "Find datasets for sentiment analysis in English"

1. dataset_search(query="sentiment", tags=["language:en", "task_categories:text-classification"], sort="downloads")
2. hub_repo_details(repo_ids=["top-dataset-id"], repo_type="dataset", include_readme=true)
```

### Discover AI Tools (MCP Spaces)

```
User: "Find a tool that can remove image backgrounds"

1. space_search(query="background removal", mcp=true)
2. dynamic_space(operation="view_parameters", space_name="result-space-id")
3. dynamic_space(operation="invoke", space_name="result-space-id", parameters="{...}")
```

### Generate Images

```
User: "Create an image of a robot reading a book"

1. dynamic_space(operation="discover")  # See available tasks
2. gr1_flux1_schnell_infer(prompt="a robot sitting in a library reading a book, warm lighting, detailed")
```

### Research a Topic

```
User: "What are the latest papers on RLHF?"

1. paper_search(query="reinforcement learning from human feedback", results_limit=10)
2. hub_repo_details(repo_ids=["paper-linked-model"], include_readme=true)  # If paper links to models
```

### Learn How to Use a Library

```
User: "How do I fine-tune with LoRA using PEFT?"

1. hf_doc_search(query="LoRA fine-tuning", product="peft")
2. hf_doc_fetch(doc_url="https://huggingface.co/docs/peft/...")
```

### Run a Quick GPU Job

```
User: "Run this Python script on a GPU"

hf_jobs(operation="uv", args={
  "script": "# /// script\n# dependencies = [\"torch\"]\n# ///\nimport torch\nprint(torch.cuda.is_available())",
  "flavor": "t4-small"
})
```

### Train a Model on Cloud GPU

```
User: "Run my training script on an A10G"

hf_jobs(operation="run", args={
  "image": "pytorch/pytorch:2.5.1-cuda12.4-cudnn9-runtime",
  "command": ["/bin/sh", "-lc", "pip install transformers trl && python train.py"],
  "flavor": "a10g-small",
  "secrets": {"HF_TOKEN": "$HF_TOKEN"}
})
```

### Check Job Status

```
User: "What's happening with my training job?"

1. hf_jobs(operation="ps")
2. hf_jobs(operation="logs", args={"job_id": "job-xxxxx"})
```

### Explore What's Trending

```
User: "What models are trending right now?"

model_search(sort="trendingScore", limit=20)
```

### Get Model Card Details

```
User: "Tell me about Mistral-7B"

hub_repo_details(repo_ids=["mistralai/Mistral-7B-v0.1"], include_readme=true)
```

### Find Quantized Models

```
User: "Find GGUF versions of Llama 3"

model_search(query="Llama 3 GGUF", sort="downloads", limit=10)
```

### Use a Gradio Space as a Tool

```
User: "Transcribe this audio file"

1. space_search(query="speech to text transcription", mcp=true)
2. dynamic_space(operation="view_parameters", space_name="openai/whisper")
3. dynamic_space(operation="invoke", space_name="openai/whisper", parameters="{\"audio\": \"...\"}")
```

### Schedule Recurring Jobs

```
User: "Run this data sync every day at midnight"

hf_jobs(operation="scheduled uv", args={
  "script": "...",
  "cron": "0 0 * * *",
  "flavor": "cpu-basic"
})
```

## Tool Selection Guide

| Goal | Tool |
|------|------|
| Find models | `model_search` |
| Find datasets | `dataset_search` |
| Find Spaces/apps | `space_search` |
| Find papers | `paper_search` |
| Get repo README/details | `hub_repo_details` |
| Learn library usage | `hf_doc_search` → `hf_doc_fetch` |
| Run code on GPU/CPU | `hf_jobs` |
| Use Gradio apps as tools | `dynamic_space` |
| Generate images | `gr1_flux1_schnell_infer` or `dynamic_space` |
| Check auth | `hf_whoami` |

## Tips

- Use `sort="trendingScore"` to find what's popular now
- Use `sort="downloads"` to find battle-tested options
- Set `mcp=true` in `space_search` to find Spaces usable as tools
- Use `include_readme=true` in `hub_repo_details` for full model/dataset documentation
- For jobs accessing private repos, always include `secrets: {"HF_TOKEN": "$HF_TOKEN"}`
- Use `dynamic_space(operation="discover")` to see all available Space-based tasks

More from huggingface/skills

SkillDescription
hf-cliHugging Face Hub CLI (`hf`) for downloading, uploading, and managing models, datasets, spaces, buckets, repos, papers, jobs, and more on the Hugging Face Hub. Use when: handling authentication; managing local cache; managing Hugging Face Buckets; running or scheduling jobs on Hugging Face infrastructure; managing Hugging Face repos; discussions and pull requests; browsing models, datasets and spaces; reading, searching, or browsing academic papers; managing collections; querying datasets; configuring spaces; setting up webhooks; or deploying and managing HF Inference Endpoints. Make sure to use this skill whenever the user mentions 'hf', 'huggingface', 'Hugging Face', 'huggingface-cli', or 'hugging face cli', or wants to do anything related to the Hugging Face ecosystem and to AI and ML in general. Also use for cloud storage needs like training checkpoints, data pipelines, or agent traces. Use even if the user doesn't explicitly ask for a CLI command. Replaces the deprecated `huggingface-cli`.
hf-cloud-aws-context-discoveryDiscover the user's local AWS context (active profile, region, account ID, caller identity) at the start of any AWS task. Use this skill before any other AWS work — deploying to SageMaker, creating resources, calling AWS APIs, or anything that touches an AWS account. Use it especially when the user has not specified a region or profile explicitly, when they say things like "use my AWS account", "deploy to AWS", "use my profile", or when about to make any AWS CLI or SDK call. Never guess the region or account ID — always use this skill to read it from the local configuration first.
hf-cloud-python-env-setupSet up an isolated Python environment for SageMaker / AWS work, with the right Python version and current boto3. Use this skill whenever Python code will be executed for a SageMaker deployment, training job, or any AWS automation — including when about to run `pip install`, when about to invoke `boto3`, when creating or activating a virtualenv, or when the user asks to "set up the environment". Never use system Python and never `pip install` into it. Always isolate. This skill prevents the most common failure modes: wrong Python version, dependency conflicts, and stale SDKs.
hf-cloud-sagemaker-deployment-plannerPlan and coordinate the deployment of a model to Amazon SageMaker AI. Use this skill whenever the user wants to deploy, host, serve, or expose a model on SageMaker or AWS — including phrases like "deploy a model", "host this LLM on AWS", "serve this embedding model", "deploy a reranker", "deploy a text-to-image / diffusion model", "host this for async inference", "create an endpoint", "serve my fine-tuned model", or any request that involves making a model available for inference on AWS. Use this even when the user is vague (e.g. "I just want to get this running on AWS, you figure it out"). Works for text-generation LLMs, embedding models, rerankers, classifiers, text-to-image / diffusion models — picks the right serving stack and chooses between real-time and async inference. This is the entry-point skill for SageMaker deployment work — it asks clarifying questions, picks a deployment pathway, and coordinates the other deployment skills.
hf-cloud-sagemaker-iam-preflightEnsure a usable SageMaker execution role exists before deploying or training. Use this skill whenever about to create a SageMaker endpoint, model, training job, or any resource that requires an execution role. Use it especially when the user has not provided a role ARN explicitly, when scripts are about to call `iam:CreateRole`, or when an AccessDenied error mentions an IAM action. Never blindly call `iam:CreateRole` — always check for existing roles first. This skill prevents the most common SageMaker deployment failure: trying to create IAM resources from an SSO principal that has no IAM write permissions.
hf-cloud-sagemaker-production-defaultsCreate a SageMaker endpoint (real-time, real-time scale-to-zero, or async) with autoscaling, CloudWatch alarms, and tagging enabled by default. Use this skill whenever about to create a SageMaker endpoint, write deployment code that calls `create_endpoint`, or finalize a deployment after the image URI and IAM role are known. Provides deploy.py for real-time endpoints, deploy_ic.py for real-time endpoints that scale to zero instances via inference components, and deploy_async.py for async endpoints (also scale-to-zero). This is the last step in the SageMaker deployment workflow. Never generate a bare `create_endpoint` call without these defaults — endpoints without autoscaling or alarms are demos, not deployments.
hf-cloud-serving-image-selectionPick the right serving container for a SageMaker model deployment and find its current image URI. Use this skill whenever about to deploy a model to a SageMaker endpoint and an image URI needs to be chosen — including when the user says "deploy this LLM", "host this HuggingFace model", "serve this fine-tuned model", "deploy this embedding model", "host a reranker", "serve a sentence-transformers model", or when about to hardcode any container URI in deployment code. HuggingFace-curated Deep Learning Containers are ALWAYS preferred: HuggingFace vLLM (LLMs and generative rerankers), HuggingFace vLLM-Omni (multimodal), TEI (embeddings/cross-encoder rerankers), HF Inference Toolkit (other transformers). Generic images (AWS vLLM, DJL-LMI, SGLang) are used only when no HuggingFace image is compatible — never merely because they carry a newer version. Never hardcode a container URI from memory and never default to TGI. Prevents stale-image failures and wrong-region URIs.
hf-memHugging Face CLI to estimate the required memory to load Safetensors or GGUF model weights for inference from the Hugging Face Hub
huggingface-best>-
huggingface-community-evalsRun evaluations for Hugging Face Hub models using inspect-ai and lighteval on local hardware. Use for backend selection, local GPU evals, and choosing between vLLM / Transformers / accelerate. Not for HF Jobs orchestration, model-card PRs, .eval_results publication, or community-evals automation.