What needed solving.
Long-form essays, fiction, project notes, and interactive research pages need consistent discovery without forcing every article into the same template.
- 01MDX + bespoke articles
- 02Shared content registry
- 03Pages + feed + sitemap
Simplified from the implementation and project notes.
How it fits together.
MDX files carry frontmatter and renderers. A shared registry combines them with bespoke article routes, supplying indexes, feeds, and sitemap entries from one metadata source.
Build-time processing calculates reading time. Drafts remain visible locally but are omitted from production listings; canonical metadata handles cross-posts and keeps duplicate entries out of the sitemap.
TanStack routing supports the site shell and standalone research experiences. Interactive articles can own their presentation while retaining the shared publication metadata.
Building and checking it.
An ordinary essay is added as an MDX file. A more elaborate research piece gets a dedicated route and one registry entry. Both become discoverable through the same publishing path.
This keeps content authoring separate from page plumbing while leaving room for simulations, diagrams, and other custom explanations.
This is a file-based personal publishing workflow rather than a multi-author CMS. The article-specific visual tools have their own assumptions and should be read alongside the text.
Source notes
Based on repository code and project records reviewed on October 30, 2026. Historical checks are described as recorded, not rerun for this article. Screenshots, renders, and mockups are labeled individually.
aspera-protocol/README.mdaspera-protocol/src/data/content.tsaspera-protocol/package.json