hugging-science
Discovers and evaluates scientific datasets, models, methodology posts
by k-dense-aiFree
Description
hugging-science is an agent skill for finding and evaluating scientific datasets, models, methodology posts and Spaces through the Hugging Science catalogue. Hugging Science is a curated index of scientific datasets, models, blog posts and interactive demos for machine learning researchers, covering 17 scientific domains. The skill uses two related sources. The catalogue at huggingscience.co is a static, readable index with compact and full text files, and per-domain pages written in markdown. The hugging-science organisation on Hugging Face holds community-submitted datasets, models and a changing set of Gradio, Docker and static Spaces, and not every listing offers an inference API. The skill is clear that curation does not establish quality, openness or whether something runs. You should verify each resource's author documentation and its scientific suitability. It also points out that the resources live on the wider Hugging Face Hub, and that each needs its own loader. ESM2 works with Transformers, Evo2 needs its native runtime, and OpenGenome2 needs explicit file-format handling. The catalogue is for discovery, and you use each resource through its verified loader or service.…
Highlights
- Index of 17 scientific domains
- Datasets, models, posts and Spaces
- Verifies author docs and suitability
- Notes that each resource needs its own loader
- Discovery only, not a quality stamp
Ratings & Reviews
No ratings yet
Be the first to rate hugging-science.
People also visit
hugging-science FAQ
What is Hugging Science?
A curated, LLM-friendly index of scientific datasets, models, blog posts and demos for machine learning researchers, across 17 domains.
Does being listed mean a resource is good?
No. The skill says curation does not establish quality, openness or executable compatibility.
Do all listed models have an inference API?
No. The skill notes that not every listing exposes an inference API.
How do I use a listed resource?
Through its verified loader or service. Some need Transformers, some need a native runtime, and some need explicit file handling.
How do I install it?
Run npx skills add https://github.com/k-dense-ai/scientific-agent-skills --skill hugging-science.
Is it checked for security?
Its security audits show passes from Gen Agent Trust Hub and Socket, and a warning from Snyk.
What are the best hugging-science alternatives?
Popular alternatives to hugging-science include huggingface-datasets, skill-scout and interview-me. See all hugging-science alternatives