# Whitelist policy: only the homepage, contact page, policy pages and
# spotlight pages (plus the Rendering Photos they embed) may be crawled.
# Everything else (designer app, app.js, pricing data, API routes, other
# assets) is off-limits to crawlers.
User-agent: *
Allow: /$
# Kept crawlable on purpose: /index.html now 301s to / (vercel.json). Crawlers
# must be able to fetch it to see the redirect and drop the duplicate URL.
Allow: /index.html
Allow: /contact.html
Allow: /spotlight.html
Allow: /spotlight/
Allow: /learn.html
Allow: /learn/guide/
Allow: /community.html
Allow: /privacy.html
Allow: /terms.html
Allow: /Rendering%20Photos/

# MUST stay crawlable: the public pages' entire stylesheet. Google and Bing
# render a page before judging it — block the CSS and they see an unstyled
# wall of text (and flag it mobile-unfriendly). This replaced the old
# cdn.tailwindcss.com script, which was external and so never robots-gated.
Allow: /assets/tw.css

# llms.txt — the plain-Markdown site guide written for AI assistants
# (ChatGPT, Gemini, Claude, Perplexity). It returned 200 but the catch-all
# Disallow below made every compliant AI crawler skip it, so the one file we
# wrote specifically for them was the one file they could not read
# (found 2026-08-25 — fifth recurrence of the whitelist trap).
Allow: /llms.txt

# IndexNow / site-verification key file (Bing Webmaster Tools).
Allow: /71abb468be9141e38aebbf3d1f7283ae.txt

# The sitemap itself. Google/Bing fetch it under robots rules too — without
# this line the catch-all Disallow below makes GSC report "Couldn't fetch".
Allow: /sitemap.xml

# spotlight.html and the homepage trends section render their article grid by
# fetching this JSON. Google's renderer honors robots for subresources too —
# with it blocked, spotlight.html rendered as an empty shell and GSC live test
# rejected it as a Soft 404 (found 2026-08-23).
Allow: /spotlights.json

# The designer tool and its 2D/3D editor interfaces (and the JS that powers
# them) must never be crawled or scraped. These are named explicitly below
# for clarity; the catch-all Disallow at the end already blocks everything
# outside the whitelist above.
Disallow: /designer.html
Disallow: /start.html
Disallow: /settings.html
Disallow: /app.js
Disallow: /edit-mode-3d.js
Disallow: /design-helper-3d.js
Disallow: /countertop-edit.js

# ⭐⭐ 内部页面**故意一个都不在这里点名**。robots.txt 是公开可读的：在这里
# 写 `Disallow: /<路径>` 等于对所有人发布一份「这个站有哪些内部页面」的目录
# (Kim 2026-08-14「不希望这个链接可以被找到」)。下面那条 catch-all 已经把
# 白名单以外的一切都挡住了，效果完全一样，只是不再点名。内部页面另有
# <meta name="robots" content="noindex,nofollow"> 和 vercel.json 里的
# X-Robots-Tag 两道，都不依赖这里点名。
#
# ⚠️ 所以：**以后新增内部页面时，不要往这个文件里加 Disallow 行。** 加了
# 就是把它公布出去。
#
# Catch-all: block everything not explicitly allowed above.
Disallow: /

Sitemap: https://www.flipkitchen.ca/sitemap.xml