getprovenant.github.io

The public homepage for Provenant — a fast, source-faithful license, copyright, package, and SBOM scanner. Built with Astro; deployed to GitHub Pages at the org root.

Develop

npm install
npm run dev      # http://localhost:4321
npm run build    # static output in dist/
npm run preview  # serve the built dist/

Structure

src/layouts/BaseLayout.astro     Shared head (SEO/OG/JSON-LD), header, footer
src/pages/index.astro            Landing page (hero, why, teasers)
src/pages/compare/index.astro    Provenant vs ScanCode — speed + annotated quality comparison
src/pages/blog/index.astro       Blog index
src/pages/blog/[...id].astro     Renders a post from the content collection
src/content/blog/*.md            Blog posts (add a file here to publish a post)
src/content.config.ts            Blog collection schema
src/styles/global.css            Design system (tokens, both themes)
public/assets/                   Demo GIF + benchmark chart (source: provenant/docs)

Adding a blog post

Drop a Markdown file in src/content/blog/ with frontmatter (title, description, kicker, standfirst). It appears on /blog/ and gets its own page automatically.

Deploy

.github/workflows/deploy.yml builds with Astro and publishes to GitHub Pages. It is manual-only (workflow_dispatch) for now. At launch: add a push trigger on main, make the repo public, and set Settings → Pages → Source: GitHub Actions.

Keeping figures in sync with the repo

The speed figures (hero + /compare/) and the benchmark chart / demo GIF are generated artifacts of the flagship repo. They live in src/data/benchmarks.json and public/assets/, and are refreshed by:

npm run sync            # reads ../provenant (a sibling checkout) by default
npm run sync -- /path/to/provenant

The script parses docs/BENCHMARKS.md for the headline numbers and copies docs/scan-duration-vs-files.svg + docs/provenant-demo.gif. So the flow is: regenerate the artifacts in the flagship repo → npm run sync here → commit. (A CI step could run this automatically on release; for now it's manual.)

The per-repo comparison specifics on /compare/ (curl / Redis / tokio deltas + snippets) come from ad-hoc compare-outputs runs, not a standing artifact, so they stay hand-maintained and are refreshed on each stable release.

Maintenance notes

• Keep ScanCode out of the landing-page hero/branding; it appears only as factual comparison on /compare/ and as lineage in the blog. A single non-affiliation/trademark note lives on /compare/ (nominative use — not legally required elsewhere).