Skip to content
Get packs

Hugging Science Catalog Finder

Find and vet scientific datasets, models, blogs and Spaces from the Hugging Science catalog

Source Download skill
Repository
K-Dense-AI/scientific-agent-skills
Stars
55,780
Licence
MIT
Files
8 files

What it does

Teaches the agent to search a curated index of scientific machine learning resources (biology, chemistry, climate, physics and more) and pick a suitable dataset, model or demo for a task. The agent fetches catalog pages with a bundled Python script, compares options, and follows reference notes on how to load each resource. It also reminds the agent to pin exact versions and check licences, since curation does not guarantee quality or openness.

Great for

  • Researchers looking for a starting model or dataset in a specific scientific field
  • ML engineers who need to compare candidate scientific models before fine-tuning
  • Students exploring what open scientific ML resources exist

Skip it if

  • Generic ML tasks such as recommendation systems or chatbots, which the skill itself says are out of scope
  • Production use without verifying each resource's licence and model card, because catalog listing does not establish quality or openness
  • Anyone expecting the skill to run scientific models for them; it only guides discovery and usage

Try asking

  • “Find a protein language model suitable for a small laptop experiment in biology”
  • “Compare genomics datasets for fine-tuning a DNA sequence classifier”
  • “Show me materials-science models on the Hugging Science catalog and explain the tradeoffs”
  • “Find a Gradio demo for protein binder design and tell me how to call it”
Step-by-step guide: how to use Hugging Science Catalog FinderInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 8 original files, the MIT licence and an attribution note.

    hugging-science.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip hugging-science.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip hugging-science.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Ask for a specific domain or model name so the agent fetches the matching topic file rather than the full catalog
  • Ask the agent to record the exact repository commit, dataset config and licence before using a resource in research
  • Review any gated or large model's memory requirements and licence on its Hugging Face page before running it

What's inside

  • references/flagship-resources.md
  • references/review.md
  • references/topics-and-slugs.md
  • references/using-datasets.md
  • references/using-models.md
  • references/using-spaces.md
  • scripts/fetch_catalog.py
  • SKILL.md

Redistributed unmodified from K-Dense-AI/scientific-agent-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.