Crate template

A self-describing repository. Drop files into data/, and an RO-Crate (ro-crate-metadata.json) is kept in sync as the single source of truth. Edit its metadata with the four buttons below — no tools required, just GitHub.

✏️ Edit this crate

     

• Configure the dataset — title, description, license, creators for the whole dataset (the root).
• Add a reference — a remote thing the dataset points to: a person, publication, the software you used, a funder, or large data hosted elsewhere — by DOI / ORCID / ROR / URL.
• Edit a data entity — describe one local file or folder (add files to data/ first).
• Tag website content — mark a file with the role it plays on the page (graphical abstract, figure, setup diagram…), so the website knows what to show where.

Each button opens a pre-filled issue form; on submit, a GitHub Action writes your answers into the crate and comments back the equivalent crate … command.

How it works

• Add data by just adding files (start with data/). Every push, crate build mirrors the filesystem into the crate as data entities — automatically.
• The crate is the source of truth. There's no seed file; the four forms (and the crate CLI, or Crate-O) edit it in place.
• Enrich resolves the DOIs / ORCIDs you reference (names, titles, authors…), best-effort.
• The published page is a generated projection of the crate — don't hand-edit the crate's derived layer (file manifest / git provenance); it's refreshed on every push.

Local use (optional)

# install pixi once (https://pixi.sh), then — from the repo root:
pixi shell           # locked env from pixi.toml/pixi.lock (same one CI uses)
crate seed --name "My dataset" --license CC-BY-4.0 --author 0000-0002-1825-0097
crate build .        # scan repo -> crate
crate enrich .       # resolve ORCIDs / DOIs
crate validate .     # check it (add --strict to gate on catalogue-readiness)
crate render . -o _site