Repository navigation
feat: add the blog, its pages and its header links - #92
Merged
Merged
Conversation
The blog follows the changelog: a `blog` collection of one post per file in content/blog/, a listing at /blog and a page per post, both prerendered in the Docus shell, and a "Blog" link after Changelog in both headers. Each post writes its own head tags and a BlogPosting, rather than the Article Docus's useSeo() would add, and has a markdown copy at /raw/blog/<slug>.md, served from its source with a canonical Link header. The sitemap, llms.txt and check-seo cover the blog, and a content gate checks every token name in a post against the Design Tokens page. The first post is a stub, which its writer replaces. Co-authored-by: multica-agent <github@multica.ai>
🚀 Deploying Preview to Cloudflare 🚀Preview URL: https://blog-uxf-871.opencomponents.dev, https://blog-uxf-871-opencomponents.blue-water-0397.workers.dev (commit 481a337)This URL reflects your latest Preview deploymentPreview Deployments by commit
|
4 tasks done
alexgrozav
added a commit
that referenced
this pull request
Nov 19, 2026
…94) ## What changed opencomponents.dev gets a blog at `/blog`, and its first post is "How to name design tokens". People search for that question, and the answer is on the Design Tokens page, spread across a reference. The post walks through it step by step and sends readers back to the page. Posts aren't part of the standard, so the MCP server doesn't read them. The branch brings together three pull requests: - #92 adds the blog. It's a Nuxt Content collection built like the changelog, with a listing at `/blog` and a page per post laid out like a long-form article: a large headline, the standfirst, the author, date and reading time, then a narrow column of body text. Both headers link to it after Changelog. Each post has `BlogPosting` structured data, a place in the sitemap and `llms.txt`, and a Markdown copy at `/raw/blog/<slug>.md`, served from its source so its tables stay tables. `pnpm check-seo` now checks all of this. A content check in `server/lib/blog.test.ts` reads every token name in a post against the pattern on the Design Tokens page, so the two can't drift apart. `.github/CONTRIBUTING.md` and `AGENTS.md` describe the blog. - #93 writes the post. Its first sentence answers the question on its own. Six steps follow, each with its reason, its rule ID and a link to the section of the Design Tokens page it comes from, then worked examples. The Design Tokens page gains one sentence that links to the post, and nothing else on it changes. - #95 rewrites the post in the voice of the docs. One example, a theme and a tooltip, is renamed a step at a time, and each step gives its reason. Links to the Design Tokens page sit where a reader would want the detail, in place of a rule ID after every claim. It's shorter, and only the post's file changes. Merging this publishes the blog and its first post. ## Checklist - [x] The title follows [Conventional Commits](https://www.conventionalcommits.org), as in `feat(docs): add the focus guideline`. - [x] `pnpm check-types` passes. (Run on each part, #92 and #93, along with `pnpm test`, `pnpm check-reference`, `pnpm build`, `pnpm check-seo`, `pnpm build:mcp` and `pnpm check-mcp`.) - [x] Docs changes: the pages read well on `pnpm dev`. (Checked on the built site, in screenshots of the listing and the post at 375px and 1280px, in light and dark mode.) - [x] Homepage changes: `pnpm build`, then `pnpm lighthouse`, still scores 100 in every category. (The homepage gains only the Blog link. Lighthouse passed on each part.) --------- Co-authored-by: multica-agent <github@multica.ai>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What changed
This adds a blog to opencomponents.dev, built the way the changelog is, so we have somewhere to write about building components beyond what the standard says. It targets the
blog-uxf-803integration branch, notmain, and its first post is a stub until the real one is written. Here's what it adds:blogcollection incontent.config.tsholds one post per file incontent/blog/./bloglists them newest first, a row each with the date, title, standfirst, author and reading time, and/blog/<slug>shows each one: a large headline, the standfirst, a byline and the body in a column of about 32em, some 65 characters a line. Both are prerendered, in the Docus shell, with the site's own colours, fonts and components.aria-current="page"on the blog's pages, and lists it in its menu below 768px. On the homepage it hides below 640px with Changelog.BlogPosting, without theArticleDocus'suseSeo()would add, and an Open Graph image. The sitemap lists the blog and its posts,llms.txtlists every post, and each post has a markdown copy at/raw/blog/<slug>.md, served from its source like the docs pages' copies, with a canonicalLinkheader.pnpm check-seonow checks all of this for the blog too.server/lib/blog.test.tschecks every token name in a post's code against the pattern on the Design Tokens page, read from the page itself, apart from the names a post lists incounterExamples. It checks the syntax only, so a reviewer still checks that each counter-example is really wrong..github/CONTRIBUTING.md, its lines in "Keeping 100s as the site grows", andcontent/blog/inAGENTS.md.Checklist
feat(docs): add the focus guideline.pnpm check-typespasses.pnpm dev. (Checked on the prerendereddist/instead, in light and dark mode at 375px and 1280px.)pnpm build, thenpnpm lighthouse, still scores 100 in every category.