Shaun Porwal's Site Repository

Source for shaunporwal.com — plain static HTML/CSS, no build step, no framework.

Quick Start

make dev    # hot-reload dev server (site/)
make test   # run tests
make sync   # regenerate nav/SEO/sitemap/listing by hand

Everything published lives under site/; everything else (templates/, scripts/, tests/, docs/) is repo tooling, not site content. Pushing to main publishes site/ to gh-pages via .github/workflows/publish.yaml — no build step, the committed HTML is what ships.

Enable the pre-commit hook once per clone (keeps nav/SEO/sitemap/listing in sync automatically):

git config core.hooksPath .githooks

See docs/development.md for adding a new post, what each script does, and full project structure.

Conventions (for agents/future contributors)

• site/ is the only published output. Content never goes anywhere else; tooling (scripts/, templates/, data/, tests/, docs/) never gets published.
• Generated content lives between <!-- X:START -->...<!-- X:END --> markers (nav, SEO meta, blog listing, resume body). Never hand-edit inside a marker block — edit the source it's generated from (templates/nav.html, data/resume.json, a post's own content) and run make sync. If you're editing inside a marker block, stop and find the generator instead.
• No framework, no build step, no npm/Node dependency committed to the repo (scripts/dev.sh uses npx for local hot reload only — nothing installed into the repo itself).
• Every script in scripts/ is small pure functions (build_block, inject, sync, etc.) plus a thin main() — not top-level script code — so it's importable and testable. Add a matching test in tests/ for any new script or behavior change, and run make test before committing.
• Adding a post = one new site/posts/<slug>/index.html (+ optional .draft). Nothing else needs touching — the pre-commit hook regenerates nav/SEO/listing/sitemap.
• Resume content changes go in data/resume.json, never directly in site/resume.html.
• Name, base URL, and social/contact links (GitHub, X, LinkedIn, email, cal.com) live in data/site.json — never hardcode a new copy of one of these elsewhere; read it via scripts/lib_site.py instead.