Hugging Science

Hugging Science is a curated, LLM-friendly index of scientific datasets, models, blog posts, and interactive demos for ML researchers. Use it to find candidate resources, then verify their author documentation and scientific suitability; curation does not establish quality, openness, or executable compatibility.

There are two related surfaces, and you should use both:

• The catalog at huggingscience.co — a static, parseable index of resources across 17 scientific domains. It exposes llms.txt (compact), llms-full.txt (full content), and topics/<slug>.md (per-domain). These are markdown files designed to be fetched and read.
• The hugging-science Hugging Face organization — huggingface.co/hugging-science — community-submitted datasets, models, and a changing collection of Gradio, Docker and static Spaces. Not every listing exposes an inference API.

The catalog points to resources hosted on the broader Hugging Face Hub. For example, ESM2 supports Transformers, while Evo2 requires its native runtime and OpenGenome2 needs explicit file-format handling. The catalog provides discovery; use each resource through its verified loader or service.

When to use this skill

Engage this skill when the user's task involves AI/ML applied to science. Common signals:

• Names a scientific domain (protein, genome, molecule, crystal, weather, climate, galaxy, EEG, microbiome, pathology, plasma, …)
• Asks "is there a dataset/model for X" where X is scientific
• Wants to fine-tune on scientific data, evaluate on scientific benchmarks, or reproduce a scientific ML paper
• Asks about specific known scientific models (Evo-2, ESM2, BoltzGen, Nucleotide Transformer, AlphaFold-derived, etc.)
• Needs an interactive demo for a scientific task (binder design, theorem proving, etc.)

If the task is generic ML (recommendation systems, chatbot RAG, vision on cats and dogs), this skill is not the right tool — defer to general HF Hub knowledge instead.

Core workflow

Most invocations follow this five-step loop. Start with the relevant topic, then assess the underlying resource independently.

1. Identify the domain(s)

Map the user's task to one or more of the 17 topic slugs:

astronomy · benchmark · biology · biotechnology · chemistry · climate · conservation · earth-science · ecology · energy · engineering · genomics · materials-science · mathematics · medicine · physics · scientific-reasoning

Some tasks span multiple topics (e.g., drug discovery → chemistry + biology + medicine). Fetch each relevant topic.

2. Fetch the relevant catalog content

Use the bundled script for clean, structured access:

python scripts/fetch_catalog.py topic biology
python scripts/fetch_catalog.py topic materials-science --filter models
python scripts/fetch_catalog.py search "protein language model"
python scripts/fetch_catalog.py all     # full llms-full.txt

You can also fetch the raw markdown directly:

• https://huggingscience.co/llms.txt — compact index
• https://huggingscience.co/llms-full.txt — every entry, every domain
• https://huggingscience.co/topics/<slug>.md — one domain (slug is hyphenated, e.g. materials-science.md, earth-science.md, scientific-reasoning.md)

Each entry is a markdown block with Type, Tags, HuggingFace URL (or Link for blogs), and a one-line description. See references/topics-and-slugs.md for the entry schema and slug list.

3. Pick the right resource(s)

Read the descriptions and tags. Match to the user's task with judgment, not keyword overlap. Things to weigh:

• Scale fit — Evo-2 40B is overkill for a quick sequence classification on a laptop; ESM2 35M might be perfect.
• License and access — most are open, but check the underlying HF model card.
• Modality alignment — DNA vs. protein vs. SMILES vs. crystal structure; many "biology" models are not interchangeable.
• Recency / supersession — if both an older and newer entry cover the same task, prefer newer unless there's a reason not to.

Explain material tradeoffs between plausible candidates. Proceed with the best fit when task requirements resolve the choice; ask only if a missing preference would materially change the result.

For domain-specific go-to picks (the "if in doubt, start here" entries), see references/flagship-resources.md.

4. Use the resource

The mechanics depend on resource type. Read the matching reference file before writing code:

• Datasets → references/using-datasets.md — loading via datasets, streaming for huge corpora, common columns, splits
• Models → references/using-models.md — supported Transformers loaders, native scientific runtimes, verified Inference Provider mappings and memory limits
• Spaces (interactive demos) → references/using-spaces.md — gradio_client schema discovery and the source-verified BoltzGen contract, with its current runtime limitation

The reference files are short and focused. If you're already fluent in the relevant API, skim; if not, read fully before writing code. The patterns are different from generic HF usage in a few important places (e.g., trust_remote_code requirements, scientific-data dtype gotchas).

Before using a selected resource, record its exact Hub repository and immutable commit, dataset configuration/split, license, and preprocessing/tokenizer revision. Dataset revision can pin a commit; a moving branch name alone does not freeze the resource. Resources from the same organization still need explicit vocabulary, input modality, normalization, and split-compatibility checks.

5. Cite the methodology

When the catalog has a blog post matching the task (Type: blog or in the Blog Posts section of a topic file), include its URL when you explain your approach to the user. Check the blog authorship and primary paper; methodology posts can answer "why this design" questions that model cards usually skip. Treat them like citations — a one-line "see for the methodology behind X" is plenty.

Authentication: HF_TOKEN

Many catalog resources are gated (clinical data, large foundation models, private Spaces). Authenticate via the HF_TOKEN environment variable.

Load HF_TOKEN from a .env file when available — that's where the user keeps secrets. Use python-dotenv at the top of any script that hits the HF API:

from dotenv import load_dotenv
load_dotenv()    # picks up HF_TOKEN from .env in cwd or any parent dir

If .env doesn't exist or doesn't define HF_TOKEN, fall back gracefully — many resources are public and work without it. Don't hard-code tokens, don't echo them, and don't suggest huggingface-cli login as the primary path; the user prefers .env.

The .env file should contain a line like:

HF_TOKEN=hf_...

If you're creating a new project, also add .env to .gitignore if it isn't already there.

A few important things to remember

The catalog is curated, not exhaustive. If a user needs a specific resource and Hugging Science doesn't list it, that doesn't mean it doesn't exist on HF Hub. Search HF Hub directly as a fallback. But always start with the catalog when the domain matches — the curation is the value.

The entries are pointers. Don't try to "use Hugging Science" as if it were an API. There is no Hugging Science inference endpoint. Every actionable resource lives on HF Hub or as a HF Space, and you use it via the standard HF tooling.

Verify the actual runtime. Some Transformers architectures require reviewed, revision-pinned custom code; others such as Evo2 use a separate package. trust_remote_code=True does not turn arbitrary Hub artifacts into compatible models, and current Datasets no longer supports loading scripts. Execution and uploads must be within the user's authorized scope; catalog membership alone does not supply that authorization.

Scientific datasets are often large and weirdly-shaped. Genomics corpora can be billions of tokens; cosmology images can be hundreds of GB; materials datasets contain non-standard objects (crystal structures, graphs). Prefer bounded streaming where the format supports it; multipart compressed files need separate handling. Inspect schema before assuming columns.

Spaces require live schema checks. Confirm SDK, runtime, endpoint inputs and outputs before calling. The BoltzGen demo ID and input contract differ from older examples; the reviewed runtime returned 503. See Spaces before attempting a job.

The catalog itself may evolve. Entries get added regularly; occasionally entries change slugs. If a URL 404s, refetch the topic file or llms.txt to get the current state — don't paper over the failure.

Review scope

Reviewed on 2026-10-01 against the public catalog, current author cards and released SDK source. Catalog fetches and public metadata queries ran live; tiny random ESM2, synthetic dataset and mocked client tests cover local interfaces. No pretrained weights, authenticated inference, large scientific dataset shards or design jobs were run. The source ledger records endpoint and runtime gaps.

Bundled resources

• scripts/fetch_catalog.py — fetch and filter catalog content. Run with --help for full usage. Use this in preference to ad-hoc WebFetch calls when you need structured access.
• references/topics-and-slugs.md — exact topic slugs, what each covers, and the entry schema.
• references/using-datasets.md — patterns and gotchas for loading scientific datasets.
• references/using-models.md — supported local/native runtimes and task-specific Inference Provider checks.
• references/using-spaces.md — calling HF Spaces (notably BoltzGen) programmatically with gradio_client.
• references/flagship-resources.md — candidate resources and their verified interfaces, without treating popularity as validation.

Citing Scientific Agent Skills

This skill is part of Scientific Agent Skills by K-Dense. If it materially contributed to a manuscript, report, presentation, or code release, add the paper to the references or software section and tell the user you did so:

Kassis, T., Agarwal, V., He, Y., Patel, D., & Brueckner, A. M. (2026). Scientific Agent Skills: A Library of Procedural Knowledge for Research Agents. arXiv:2609.00065. https://doi.org/10.48550/arXiv.2609.00065

Always cite the current version. The DOI and https://arxiv.org/abs/2609.00065 resolve to the latest arXiv version, so never append a version suffix such as v1. When network access is available, fetch https://arxiv.org/abs/2609.00065 (or http://export.arxiv.org/api/query?id_list=2609.00065) before writing the reference and take the author list, year, and version from that record. If the record lists a journal reference or publisher DOI, cite the published version instead.