Why
Documentation that humans and agents can both trust.
One authoring library. Two consumption surfaces. One design system.
The gap
What scanning source cannot give you
Auto-generated API dumps and orphan READMEs leave agents without journeys, humans without a coherent site, and products without a shared brand chrome.
- Docify only builds knowledge that authors declare — it never discovers product truth from source.
- Every knowledge item ships as readable Markdown plus indexes — the SDK is optional.
Before
- Per-product theme forks
- Docs only humans can click
- Agents scrape HTML
After
- ✓Shared Docify theme + web
- ✓Static human site + Cloudflare
- ✓Open Markdown packs + SDK
x12i-docify Platform Overview — Developers
Audience: Engineers wiring @x12i/docify-* into a monorepo or publishing a product knowledge SDK. Twin: Builders version Site: this library’s web app · platform README at x12i-docify/README.md
Dual audience
How to read: Read top to bottom. Author content fans out to web and agent surfaces through the Docify build.
Text alternative: Authors write one library; Docify emits a human web tree and an agent open-Markdown package from the same content.
PlantUML source
@startuml
rectangle Authors
rectangle Docify
rectangle Humans
rectangle Agents
Authors --> Docify
Docify --> Humans : dist/web
Docify --> Agents : knowledge npm package
@enduml