Writing, projects, machines, and objects by RUP L.
This is the source repository for the BOUNDARY GitHub Pages site. It is currently published through the ize-studio GitHub account.
The site is a Jekyll archive for essays, fiction, reading notes, writing tools, project records, Korean writing links, and future small editions or objects.
bundle exec jekyll build
bundle exec jekyll serveThis repository is built by GitHub Actions and published to ize-studio/ize-studio.github.io.
_posts/: English archive posts. Existing post URLs use/posts/:categories/:title.html.writing/: editorial writing index.fiction/: fiction index.projects/: project record index.shop/: object and edition shelf._products/: optional product collection. Draft products are not shown._data/brunch.json: cached Brunch article metadata.scripts/fetch_brunch.py: Brunch metadata sync script.
Add Markdown files to _posts using the Jekyll filename format:
YYYY-MM-DD-title.md
Front matter:
---
layout: post
title: "Post Title"
date: 2026-06-03
category: Essays
---category and categories are both accepted:
category: Essayscategories: [Essays]Category aliases include singular, plural, and Korean names for the existing archive pages.
Add a Markdown file to _products/:
---
title: "Object Title"
category: Book
year: 2026
status: upcoming
price:
currency: KRW
cover:
purchase_url:
---Supported status values:
availablesoldoutupcomingarchivedraft
draft products are not shown on the site.
The Brunch cache is metadata only. It stores title, URL, date, excerpt, thumbnail, source, and language.
Run manually:
python scripts/test_fetch_brunch.py
python scripts/fetch_brunch.pyThe scheduled workflow is .github/workflows/sync-brunch.yml. If parsing returns zero posts or fails, the script keeps the previous _data/brunch.json cache.
.github/workflows/publish.yml builds this source repository with Jekyll and publishes the generated site to the public GitHub Pages repository.