Folio

Project-agnostic documentation engine. Aggregates markdown from remote Git repositories into a single static site.

How it works

1. Define sources in folio.config.ts (Git repos with docs directories)
2. folio stage fetches docs via sparse-clone, skipping unchanged repos
3. folio build renders everything with Astro and indexes with Pagefind

Quick start

bun install
bun run stage # fetch docs from remote repos
bun run dev # stage + start dev server
bun run build # stage + build + pagefind index

Use --force to bypass the cache and re-fetch all sources.

Configuration

// folio.config.tsexportdefault{site: {title: 'My Docs',},sources: [{name: 'my-project',url: 'https://github.com/org/repo',docsDir: 'docs',mount: '/my-project',},],};

Each source supports:

Field	Default	Description
name	required	Identifier for caching
url	required	Git remote URL
branch	main	Branch to fetch
docsDir	docs	Directory containing markdown
includeFiles	["README.md"]	Extra files from repo root
mount	required	URL path prefix

Deploy

Container stack with nginx, a rebuild cron, and a Pangolin tunnel via Newt.

cd deploy
cp .env.example .env
# fill in credentials
podman-compose up -d

Pangolin setup

Create a new site in your Pangolin dashboard for your docs domain (e.g. docs.example.com). Configure the resource target to point at the nginx container:

Setting	Value
Target host	folio-nginx
Target port	80
Protocol	http

Then create a Newt and copy the credentials into deploy/.env:

PANGOLIN_ENDPOINT=https://pangolin.example.com
NEWT_ID=your-newt-id-here
NEWT_SECRET=your-newt-secret-here
REBUILD_INTERVAL=21600

License

MPL-2.0. See LICENSE.