Synergy Tweaks website

The static site for Synergy Tweaks — home for SynergyOS, SynToolkit, docs and the team.

Plain HTML, CSS and a little JavaScript. No framework, no build step, no runtime dependencies. Drop the folder on any static host and it runs.

Pages

File	What it is
index.html	Home: what we build, facts, comparison, downloads
synergyos.html	Playbook: requirements, install steps, download, risks, FAQ, changelog
syntoolkit.html	The WinUI 3 companion app
docs.html	Documentation index
about.html	Team and contributors
privacy.html	Privacy policy
credits.html	Credits and third-party libraries

Structure

styles.css            design tokens + components
js/config.js          org links, Discord, socials, download fallbacks
js/main.js            release data, downloads, copy-SHA, reveals
data/releases.json    generated release snapshot (do not edit by hand)
data/known-issues.json
data/changelog.json   offline fallbacks
images/               logos, screenshots, profile art, icons, OG card
fonts/                Nunito + Manrope, self-hosted
tools/check.mjs       page check
tools/fetch-releases.mjs  regenerates data/releases.json
.github/workflows/    release refresh + page check
robots.txt
sitemap.xml

Local development

python -m http.server 8080

Open http://localhost:8080/. Any static server works.

After editing styles.css or js/main.js, bump the ?v= number in the HTML so browsers do not keep the old file.

Release data

Publishing a GitHub release needs no site edits. On load, js/main.js fills in version, filename, size, SHA-256 (asset digest), download count, download link and release notes straight from the GitHub API, then caches the result for an hour.

data/releases.json is a snapshot for the cases where the API is rate-limited or offline:

• .github/workflows/releases.yml refreshes it every 6 hours (or on demand) and commits only when a release actually changes.
• It carries no download counts, so the scheduled job does not commit on every run.
• When the API fails, version, size, SHA and notes still render from the snapshot; the download count hides rather than showing a wrong number.

Download buttons always resolve the newest asset. The plain href points at /releases/latest, never a version-pinned file.

Checks

node tools/check.mjs

Runs on every push and pull request via .github/workflows/check.yml. It catches dead links, missing images, broken tags, version-pinned download URLs and leftover placeholders.

Configuration

js/config.js holds Discord, GitHub, YouTube, AME Wizard and download fallbacks. Keep matching hrefs in the HTML in sync for the no-JavaScript case.

Deploying

Any static host works — GitHub Pages, Cloudflare Pages, Namecheap, plain Apache. Upload the repository root; no server or database is involved.

GitHub Pages (free for public repositories):

1. Settings → Pages
2. Source: Deploy from a branch → main → / (root)
3. Save. The site appears at https://<org>.github.io/Website/ until a custom domain is connected.

Before a custom domain

TODO-SET-DOMAIN placeholders sit in the HTML meta tags, robots.txt and sitemap.xml. Replace them with the production origin, then set absolute canonical / og:url / og:image URLs and a Sitemap: line.

Contributing

Issues and pull requests are welcome. Run node tools/check.mjs before opening a PR, and do not hand-edit data/releases.json — it is generated.

License

Website code © Synergy Tweaks. SynergyOS and SynToolkit keep their own licenses in their repositories.

Requirements

• A modern browser with JavaScript enabled for the dynamic parts (release data, scroll reveals, copy-SHA). The site still reads without JavaScript.
• Python 3 only if you use python -m http.server; any static file server works.
• Node.js 20 (the version CI uses) only for tools/check.mjs and tools/fetch-releases.mjs.
• No package manager, no dependency install, no compile step.

How the site fits together

Each page in the root is a standalone HTML document. The header, navigation and footer markup is repeated per page; styles.css holds the design tokens and all component styles.

js/config.js runs first and puts external URLs and release fallbacks on window.SITE. js/main.js then:

• writes hrefs onto elements carrying data-link (the hard-coded HTML hrefs are the no-JavaScript fallback),
• toggles the scrolled header state and closes the mobile menu on navigate,
• reveals .reveal elements with IntersectionObserver, skipping the animation when the visitor prefers reduced motion,
• copies a release SHA from the target element,
• hydrates [data-release][data-field] slots and a[data-repo] download links.

Content that does not change between releases lives in the HTML; anything tied to a GitHub release comes from the API at runtime, falling back to data/*.json. tools/ holds the two Node scripts, and .github/workflows/ runs them in CI.