Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Hugging Science Resource Finder

Discovers and applies curated Hugging Face datasets, models, and Spaces across 17 scientific domains, from protein design to climate modeling.

Data & AnalyticsIntermediate★ 33,030⑂ 3,248AI score 9/10Last updated: Aug 9, 2026

What it does

  • Fetches the curated Hugging Science catalog (llms.txt, llms-full.txt, topics/<slug>.md) to surface high-signal scientific datasets, models, methodology blogs, and Spaces.
  • Maps your task to one or more of 17 topic slugs (astronomy, biology, chemistry, climate, genomics, materials-science, medicine, physics, scientific-reasoning, …) and uses the bundled scripts/fetch_catalog.py for structured search and filtering.
  • Walks you into real usage: streaming loads with datasets, local or hosted inference with transformers / HF Inference API / Inference Providers, and gradio_client calls to Spaces such as BoltzGen.
  • Weighs candidates by scale fit, license, modality, and recency, presenting 2–3 options when the choice matters, and cites author-written methodology blogs.
  • Handles gated resources by loading HF_TOKEN from .env via python-dotenv, with rules against hard-coding or echoing tokens.

Who it's for

  • Researchers and grad students doing ML on proteins, genomes, molecules, crystals, or weather data.
  • Engineers reproducing a scientific ML paper or evaluating on scientific benchmarks.
  • Anyone lost among thousands of HF Hub repos who wants a sensible domain default.

Examples

  1. "Which model should I use for protein sequence classification?" → pulls the biology topic and compares ESM2 35M/650M vs. Evo-2 on cost and hardware fit.
  2. "Design a binder for this target protein" → calls the hosted BoltzGen Space through gradio_client instead of provisioning a GPU.
  3. "Prep a fine-tune on a multi-GB genomics corpus" → inspects the schema with streaming=True first, then scaffolds training and links the methodology blog.

Note: the catalog is curated, not exhaustive, and inclusion is not a code review. Models needing trust_remote_code=True should be approved by you first.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/K-Dense-AI/scientific-agent-skills/HEAD/skills/hugging-science/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/hugging-science folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/hugging-science/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/hugging-science ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Clone the repository in your terminal: 
    git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  2. Create the Claude Code skills directory: 
    mkdir -p ~/.claude/skills
  3. Copy just this skill folder (its scripts/ and references/ must come along): 
    cp -r scientific-agent-skills/skills/hugging-science ~/.claude/skills/
  4. Install the Python dependencies: 
    pip install datasets transformers gradio_client python-dotenv requests
  5. For gated datasets/models, create a .env file in your working directory containing HF_TOKEN=hf_..., and add .env to .gitignore.
  6. Restart Claude Code and ask something like "find me a protein language model" — the skill triggers automatically.
View source on GitHub ↗License: MIT