Design research
Run the three-window research method on any design question: /last30days (engagement — what's alive now), /last30years (survival — what's load-bearing), /last300years (endurance — what governs taste). Trigger: "research X for this design", "what's the state of the art in Y", "has Z ever worked".From its SKILL.md
npx -y skills add wjlgatech/design-anything --skill design-researchAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its file declares
Copied from the file, not written here
The file declares its own license as CC-BY-4.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
1.8 KB, 355 tokens by cl100k_base, as published. Nobody here has run it
design-research
When to use
Before betting the repo, a project, or a client on a tool, method, or principle.
What it does
- 30 days: rank by observed engagement — releases, stars, pushes, dated announcements. Freshest verified signal wins; reputation counts for nothing.
- 30 years: rank by survival — still-used + built-upon + hype-survived (the 3-probe test). Tier 🏛/🌳/🌱; under-10-years can never outrank a survivor.
- 300 years: rank by endurance with a Whig-history guard — keep the surviving kernel, flag the superseded dogma (Modulor's ruler, not his city).
- Synthesize the Lindy read: bet interfaces on survivors, hot-swap the young layer, and record the anti-portfolio (who was louder and died).
Example
The repo's own digests are the golden output: research/last30days/DIGEST.md,
research/last30years/DIGEST.md, research/last300years/DIGEST.md.
Verification (eval-with-teeth)
Every entry carries a URL; dated claims carry dates; anything unverifiable is marked UNVERIFIED; single-source promotional claims are flagged as such. A digest with unlabeled uncertainty is a failed run — no evidence ⇒ No.
Safety
Research grounds decisions; it doesn't make them. Surface trade-offs, let the human pick at genuine forks.
Cross-runtime
Needs web search in the harness; degrades to knowledge-cutoff answers with an explicit staleness warning.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.