API & Agent Discovery
This page documents the public discovery surface of help.doran.app. It is intended for crawlers, AI agents, and integrators that need to enumerate documentation and consume it in machine-friendly formats.
Discovery resources
- Linkset —
/.well-known/linkset(RFC 9264) — lists the canonical entry point for each documentation section. - API catalog —
/.well-known/api-catalog— full JSON manifest of every doc page, grouped by section, with titles, descriptions, and markdown URLs. - Service docs — every documentation section is advertised via a
Link: <…>; rel="service-doc"header on the homepage.
Fetching markdown instead of HTML
Every documentation page is available as plain markdown via two equivalent mechanisms.
Content negotiation (recommended)
Send Accept: text/markdown on any /docs/<slug> request to receive the markdown body of that page instead of the rendered HTML.
curl -H "Accept: text/markdown" https://help.doran.app/docs/account/staff-accountThe response uses Content-Type: text/markdown; charset=utf-8 and Cache-Control: private, max-age=3600. Both the HTML and markdown responses are pre-generated at build time, so the markdown endpoint adds no rendering latency.
Dedicated markdown endpoint
The same markdown body is also served at /mdx/<slug> as a stable URL. Use this when you cannot set the Accept header.
curl https://help.doran.app/mdx/account/staff-accountCanonical URL patterns
| Pattern | Description |
|---|---|
https://help.doran.app/docs/{slug} | Documentation page (HTML) |
https://help.doran.app/mdx/{slug} | Same page as markdown |
https://help.doran.app/.well-known/linkset | Discovery linkset (RFC 9264) |
https://help.doran.app/.well-known/api-catalog | Full content manifest (JSON) |
https://help.doran.app/llms-full.txt | All docs concatenated as plain text |
https://help.doran.app/sitemap.xml | Sitemap for crawlers |
Documentation sections
Each section's index page is advertised as a service-doc and is enumerated in the API catalog.
Catalog payload
The API catalog is a static JSON document regenerated on every build. Fields:
site— site name, URL, and description.discovery— pointers back to the linkset, catalog, documentation root, and URL patterns.content.counts— totals for docs and sections.content.sections— one entry per top-level section, withslug,title,url, anddocCount.content.docs— one entry per documentation page, withslug,title,description,section,url, andmarkdownUrl.