ws-artefacts-explainedlisted
Install: claude install-skill wsagency/WS-Claude-marketplace
# ws-artefacts Explained Repos
A hub MAY register one sub-repo with `role: explained` (max ONE per hub, same
validation rule as `role: docs`). It holds **generated, human-facing visual
documentation of the whole product**: self-contained HTML artefacts published
through the [ws-artefacts](https://artefacts.wsagency.io) platform behind
unguessable token links.
**Audience: the product owner + the dev team.** They should never have to read
`dev-docs/` or `openwiki/` directly — those serve AI agents. The explained repo
is the human window into the same knowledge.
**Explained is an OUTPUT, never a source of truth.** It is synthesized from the
hub's `project.yaml`, `openwiki/` (the primary derived map), the `role: docs`
repo's `dev-docs/`, per-sub-repo `dev-docs/`, and sub-repo READMEs. When it
drifts, **regenerate it — never hand-edit**. Commits go to the explained repo
itself (its own git; the hub ignores it like any sub-repo).
## Artefact HTML contract
Verified against the ws-artefacts repo (`build.mjs` consumer, `templates/`,
`clients/` examples):
- **One self-contained `.html` file per artefact**, flat at the repo root
(kebab-case filename, e.g. `acme-explained.html`). Full `<!doctype html>`
document, ALL CSS and JS inline, images as inline SVG or `data:` URIs —
"like Claude artifacts".
- **No external network at view time.** No CDN scripts, no external
stylesheets, no web fonts, no remote images. **NO mermaid** — render
diagrams as inline SVG at generation tim