DataPressr

Turn a raw data find into a clean, published dataset — the same way every time, whoever (or whichever model) does the work.

DataPressr is a set of opinionated skills (prescriptive playbooks) that carry a dataset through its whole lifecycle. /init, /validate and /push already covered the last mile — packaging and shipping. The newer skills cover everything upstream: the judgement-heavy part that used to be done ad hoc in a chat window.

The workflow

Stage	Skill	What happens
capture	`capture`	A URL or idea becomes a GitHub issue — near-zero friction, no judgement
archived	`archive`	The raw source is snapshotted with provenance (URL, date, licence)
structured	`structure`	Raw → tidy, typed CSV(s) + a real `datapackage.json` schema
—	`init` / `validate` / `push`	Scaffold, deterministic checks, publish to DataHub
enriched	`enrich`	Descriptive stats + first charts + a note of what stands out
story	`story`	A short data story: reviewed outline → charts → prose

The skills live in skills/ and install into any agent — npx skills add datasets/datapressr — not just Claude Code.

Honest status

Solid: the validate script and the shared wrangling idioms — a real test suite (npm test, 47 tests).
structure — proven and benchmarked. Run end-to-end on three real sources: co2-ppm (NOAA text file), Project Drawdown (Markdown table), and oil-prices (eight legacy .xls workbooks) — the last diffed against the published community datasets/oil-prices and found content-identical. Scored in docs/structure-benchmark.md; eight prioritised skill edits, all applied 2026-09-06.
capture / archive — in use. archive has snapshotted sources with provenance; capture files finds in an Inbox issue, or as a bead once substantive.
Three data stories: The Keeling Curve (#9) and Planetary Boundaries (#4), written by hand, and WTI Went Negative. Brent Didn't., the first written with the story skill. Each was built from a committed outline. The author's voice pass is outstanding on all three.
story and enrich — active. Both were drafted from hand-made examples and graduated after real runs: enrich on co2-ppm and oil-prices, story on three stories, with an independent review gate on the outline. Each skill bundles its guidance in references/ (skills/story/, skills/enrich/). Stories chart with Observable Plot (charting.md).
Not designed yet: monitor.

Full decision history: docs/skills-vision.md.

Roadmap to v1

Tracked in Beads (bd ready); the original roadmap issue, #14, is closed. v1 — the enrich + story skills plus a settled near-term charting approach — is complete (2026-09-18): both skills are active and proven on real runs, and charting is settled on Observable Plot for stories. Still open, not blocking: the author's voice passes. Post-v1: monitor + unattended cloud execution (#6); skill evals for the prompt-only skills; the full plan.

Pages

What to review now — the short list of things that actually need eyes
Datasets — what's been produced, with status and links
Changelog — what has shipped, newest first
Docs — the decision history and playbooks: skills roadmap, source-discovery playbook, structure benchmark, charting, and more under docs/
Data story: The Keeling Curve — the first hand-written story
Data story: Planetary Boundaries — the second hand-written story
Data story: WTI Went Negative. Brent Didn't. — the third story, the first written with the story skill

_{This site is site/ in datasets/datapressr. The official site at https://datapressr-2-rufuspollock.flowershow.me git-autosyncs from site/ on every push to main; a preview site (fl site --name datapressr-preview --yes) is for checking changes first.}