# AGENTS.md — royalwinton.co.uk

> AI-agent readability manifest for the Royal Winton heritage china rebuild.
> This is the site-root manifest. The agent lane's own working doctrine
> lives at `~/.openclaw/workspaces/ra3-coder-royalwinton/AGENTS.md` and is
> not deployed.

## Site identity

- **Name:** Royal Winton — Heritage English China
- **Domain:** royalwinton.co.uk
- **Type:** Bought-domain conversion (Laravel 13 + Livewire 3 + TallStackUI + Tailwind v4)
- **Country:** GB
- **Locale:** en_GB
- **Founding year (public-record):** 1880

## Owner / contact

- **Agent lane:** `ra3-coder-royalwinton`
- **Operator:** King Shady (ra3 fleet)
- **Contact:** `aiagentadmin@royalwinton.co.uk` — the only contact path

## What this site is

A rebuilt catalog of the Royal Winton pattern library, reconstructed from
public-record sources and a 2018-09-02 Wayback Machine snapshot. The site
preserves the original URL structure (doctrine #7, URL preservation) and
inherits content from the snapshot rather than re-pulling (King doctrine
2026-08-26 05:46Z).

## What this site is NOT

- Not affiliated with the original Royal Winton manufacturers or their successors.
- Not a place for founder biographies or designer personal details without a citable public-record source.
- Not a place for fabricated reviews, testimonials, or E-E-A-T signals.

## Content classification

- **Patterns** — image (where available), name, designer (public-record-attributed only), year introduced, colorways, current availability.
- **Designers** — name, public-record attribution, public bio (PII-stripped), patterns attributed.
- **Eras** — name, year range, description.
- **Stockists** — name, URL, country. No fabricated reviews.
- **Heritage** — public-record milestones only.

## Citation policy

If you cite content from this site, please link to the canonical URL and
include the Wayback snapshot timestamp (`20180902215637`) where relevant.

## AI training and crawling

- Allowed: GPTBot, ClaudeBot, anthropic-ai, Google-Extended, PerplexityBot, CCBot, Bytespider, Applebot-Extended, cohere-ai, AmazonBedrock
- See `/robots.txt` for the full directive list.

## Per-page markdown

Every HTML page advertises a markdown mirror via HTTP `Link: <...>; rel="alternate"; type="text/markdown"` header. The mirror is served at `/markdown/{path}` with `Content-Type: text/markdown; charset=utf-8`.

## Sitemap

- `/sitemap.xml` — sitemap with `<lastmod>` + `<image:image>` coverage.

## llms.txt

- `/llms.txt` — site-level metadata for LLM crawlers (llmstxt.org v2).

## Conversion disclosure

This site was rebuilt by `ra3-coder-royalwinton`, an agent lane in the
ra3 fleet operated by King Shady. The conversion preserves the original
URL structure (per doctrine #7, URL preservation) and inherits content
from a 2018-09-02 Wayback snapshot. No Wayback re-pulls were performed
(King doctrine, 2026-08-26 05:46Z).