mirror of https://github.com/adobe/skills.git synced 2026-09-19 08:06:09 +08:00

Files

T

Paolo Mottadelli 9cfed1145a feat(stardust): replace v1 with v0.3.0 (redesign on top of impeccable) (#111 )

v1 (four-stage greenfield: brand → briefings → wireframes → prototype) is
preserved at tag stardust--v0.1.0. v0.3.0 is a complete refactor with a
new surface (extract → direct → prototype → migrate) and a hard runtime
dependency on impeccable.

- plugins/stardust: replace contents with v0.3.0 source from paolomoz/stardust-2
  (skills, evals, README, tile.json, plugin.json)
- plugins/stardust/.claude-plugin/plugin.json: bump to 0.3.0; add repository
  and v2-relevant keywords
- plugins/stardust/skills/*/SKILL.md: add license: Apache-2.0 to frontmatter
- .claude-plugin/marketplace.json: update stardust entry with new
  description, version, category, keywords, author, homepage, repository,
  license (mirrors adobe-for-creativity entry shape)
- plugins/stardust/README.md: status bumped to v0.3.0; v1 link repointed
  to the stardust--v0.1.0 tag

Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>

2026-04-30 17:24:22 +02:00

2.1 KiB

Raw Permalink Blame History

Eval: multi-page extract with cap

Setup

Empty project; no stardust/ folder. Impeccable installed in .agents/skills/impeccable/ (hard dep verified by stardust setup).

User prompt

"Use stardust to extract https://stripe.com"

Expected behavior

The stardust:extract skill is invoked. It:

Runs the master skill setup (impeccable dep check, context loader).
Discovers the page inventory before crawling: tries sitemap.xml, then sitemap_index.xml, then robots.txt Sitemap: directives, then BFS crawl.
Applies the default page cap of 25 and shows the user the discovered list, the cut list, and asks for confirmation.
Uses Playwright (not WebFetch / curl) for per-page rendering at 1440 × 900 @ 2× DPR with networkidle + 1.5 s grace and a scroll-to-bottom pass.
Writes one stardust/current/pages/<slug>.json per crawled page with the full schema (headings, landmarks, ctas, links, media, forms, widgets, per-section style).
Runs the brand-surface extraction once on the home page and writes stardust/current/_brand-extraction.json with logo, palette (frequency-clustered), type, spacing, motifs, componentStyle, voice samples.
Locates the logo via the priority chain (inline SVG → <img> with logo-ish identifier → apple-touch-icon → og:image → favicon → synthesized placeholder) and saves it under stardust/current/assets/logo.<ext> (not icons/).
Authors stardust/current/PRODUCT.md directly using impeccable's teach.md as the format spec — does NOT invoke $impeccable teach (the current-state file is descriptive, no interview).
Authors stardust/current/DESIGN.md and DESIGN.json directly using impeccable's document.md as the format spec — does NOT invoke $impeccable document.
Updates stardust/state.json with site.originUrl, site.pageCap, site.crawled, and one pages[] entry per crawled page with status: "extracted".
Prints a summary report and recommends $stardust direct next.
Does NOT reference EDS, AEM, localhost:3000, or dev servers.

2.1 KiB Raw Permalink Blame History Unescape Escape

Eval: multi-page extract with cap

Setup

User prompt

Expected behavior

2.1 KiB

Raw Permalink Blame History