Openclaw Huggingface

Manage models, datasets, Spaces, and repositories using Hugging Face CLI (hf). Supports authentication, upload, download, Space creation, and more.

Security Scan

The skill is an instruction-only wrapper around the official Hugging Face CLI and requests only the HF_TOKEN and the hf binary that are appropriate for its stated purpose.

Purpose & Capability

Name/description match the actual requirements: the skill needs the 'hf' CLI and HF_TOKEN to operate on Hugging Face models, datasets, Spaces, and repos. Nothing requested appears unrelated to the stated functionality.

Instruction Scope

SKILL.md contains concrete hf CLI commands (auth, models, datasets, repos, spaces, upload/download). Commands reference local paths only when doing uploads/downloads, which is expected. The instructions do not ask the agent to read unrelated system files or to transmit data to endpoints outside Hugging Face CLI's normal behavior.

Install Mechanism

There is no install spec (instruction-only). That minimizes risk — the skill does not download or write code to disk. It expects the user/host to provide the official 'hf' binary.

Credentials

Only HF_TOKEN is required, which is the expected credential for interacting with the Hugging Face Hub. No unrelated secrets, keys, or config paths are requested. The docs note you can alternatively pass --token, which is consistent.

Persistence & Privilege

The skill is not always-on and does not request elevated persistence or modify other skills/configs. Autonomous model invocation remains possible (platform default) but is not combined with additional concerning privileges.

Assessment

This skill is essentially documentation for the official Hugging Face CLI. Before installing/use: (1) ensure you have the official 'hf' CLI binary installed (verify vendor/source) so commands run against the real tool; (2) provide an HF_TOKEN with least-privilege scopes needed (avoid a fully-scoped or long-lived token if not necessary); (3) be careful when running upload commands to avoid unintentionally publishing private data or large files; and (4) keep the hf CLI updated to avoid using a compromised client binary.

Hugging Face CLI Skill

Use Hugging Face Hub CLI (hf) for various operations.

Environment Variables

• HF_TOKEN: Hugging Face API Token (get from https://huggingface.co/settings/tokens)

Core Features

1. Authentication Management (hf auth)

# Check login status
hf auth whoami

# List all tokens
hf auth list

# Login
hf auth login

# Logout
hf auth logout

# Switch token
hf auth switch

2. Model Management (hf models)

# List models (supports sorting and filtering)
hf models ls --sort downloads --limit 10
hf models ls --search "llama"

# Get model info
hf models info meta-llama/Llama-3.2-1B-Instruct

3. Dataset Management (hf datasets)

# List datasets
hf datasets ls --limit 10
hf datasets ls --search "imagenet"

# Get dataset info
hf datasets info HuggingFaceFW/fineweb

4. Spaces Management (hf spaces)

# List Spaces
hf spaces ls --limit 10

# Get Space info
hf spaces info username/repo-name

# Hot-reload (experimental, for Gradio 6.1+)
hf spaces hot-reload username/repo-name app.py
hf spaces hot-reload username/repo-name -f ./local/app.py

5. Repository Management (hf repos)

# Create new repository
hf repos create my-model --type model
hf repos create my-dataset --type dataset
hf repos create my-space --type space

# Delete repository
hf repos delete username/repo-name

# Set as private
hf repos settings username/repo-name --private

# Manage branches
hf repos branch create username/repo-name feature-branch
hf repos branch delete username/repo-name feature-branch

# Manage tags
hf repos tag create username/repo-name v1.0
hf repos tag delete username/repo-name v1.0

# Move repository to another namespace
hf repos move old-namespace/my-model new-namespace/my-model

6. Download Files (hf download)

# Download entire model
hf download meta-llama/Llama-3.2-1B-Instruct

# Download specific files
hf download meta-llama/Llama-3.2-1B-Instruct config.json tokenizer.json

# Download with glob patterns
hf download meta-llama/Llama-3.2-1B-Instruct --include "*.safetensors"
hf download meta-llama/Llama-3.2-1B-Instruct --include "*.json" --exclude "*.bin"

# Download to local directory
hf download meta-llama/Llama-3.2-1B-Instruct --local-dir ./models/llama

# Download dataset
hf download HuggingFaceM4/FineVision --repo-type dataset

7. Upload Files (hf upload)

# Upload entire directory
hf upload my-cool-model . .

# Upload single file
hf upload username/my-model ./models/model.safetensors

# Upload to dataset
hf upload username/my-dataset ./data /train --repo-type dataset

# With commit message
hf upload username/my-model ./models . --commit-message="Epoch 34/50" --commit-description="Val accuracy: 68%"

# Create Pull Request
hf upload bigcode/the-stack . . --repo-type dataset --create-pr

# Create private repository
hf upload username/my-private-model . . --private

8. Collection Management (hf collections)

# Create collection
hf collections create "My Models"

# Add item to collection
hf collections add-item username/my-collection moonshotai/kimi-k2 model

# List collections
hf collections ls

# Get collection info
hf collections info username/my-collection

# Update collection
hf collections update username/my-collection --title "New Title"

# Update collection item
hf collections update-item username/my-collection ITEM_OBJECT_ID --note "Updated note"

# Delete item
hf collections delete-item username/my-collection ITEM_OBJECT_ID

# Delete collection
hf collections delete username/my-collection

Usage Examples

Example 1: Download and Upload Model

# Download model
hf download meta-llama/Llama-3.2-1B-Instruct --local-dir ./llama-model

# Upload to your repository
hf upload username/my-llama ./llama-model .

Example 2: Manage Space

# Create Space
hf repos create my-app --type space

# Upload code
hf upload username/my-app ./app.py

# Hot-reload for development
hf spaces hot-reload username/my-app app.py

Example 3: Batch Operations

# Download all safetensors files
hf download meta-llama/Llama-3.2-1B-Instruct --include "*.safetensors"

# Upload and create PR
hf upload username/model . . --create-pr --commit-message="Update model"

Notes

1. Token Management: Ensure HF_TOKEN environment variable is set, or use --token parameter
2. Large File Upload: For large folders, consider using hf upload-large-folder
3. Space Hot-Reload: Only works with Gradio 6.1+, experimental feature
4. Free Space Limits:
   • Free fixed vCPU: 2
   • RAM: 16GB
   • No persistent storage (use external storage or HF Datasets)

Resources

• Hugging Face CLI Documentation
• Hugging Face Token Settings
• Hugging Face Spaces