User-agent: *
Allow: /
Disallow: /app.php
Disallow: /account.php
Disallow: /decks.php
Disallow: /api/
Allow: /api/tour.php
Allow: /api/demo.php
Disallow: /go/

# The *-playground.php files are NOT listed here on purpose. They already carry
# <meta name="robots" content="noindex, nofollow">, and a crawler has to fetch a
# page to read that tag — blocking them here would leave the noindex unread and
# the urls eligible to appear as bare links. A noindex page must stay crawlable.

# AI crawlers are allowed ON PURPOSE, training crawlers included.
#
# The bet: a learner asking an assistant "where should I start with Chinese?" is
# the same person a search result would have reached, and an assistant can only
# name a source it was allowed to read. The dictionary, the HSK word lists, the
# curriculum order and every guide are the parts worth being quoted on, so they
# are open. What is sold is the lesson loop and the scheduler, not the facts.
#
# The bet is paying: Bing's AI Performance report showed the first citations on
# 2026-08-09 and 75 in the 30 days to 2026-08-16, 67 of them on /hsk-levels.
#
# `User-agent: *` already permits these. They are named because several honour
# only a directive addressed to them by name, and because an explicit allowlist
# is a record of intent rather than an oversight.
#
# Checked 2026-07-31: graphchinese.com is NOT proxied through Cloudflare (DNS
# resolves straight to the origin, no cf-ray header), so nothing in front of this
# file is filtering crawlers — GPTBot, ClaudeBot and PerplexityBot all get 200
# from the origin. If the apex is ever put behind a proxy, re-check there first:
# most bot-management defaults block AI crawlers regardless of what this says.

User-agent: GPTBot
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: ClaudeBot
User-agent: Claude-User
User-agent: Claude-SearchBot
User-agent: Claude-Web
User-agent: anthropic-ai
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Google-Extended
User-agent: GoogleOther
User-agent: Google-CloudVertexBot
User-agent: Applebot
User-agent: Applebot-Extended
User-agent: Bingbot
User-agent: CCBot
User-agent: meta-externalagent
User-agent: Meta-ExternalFetcher
User-agent: FacebookBot
User-agent: Amazonbot
User-agent: DuckAssistBot
User-agent: MistralAI-User
User-agent: cohere-ai
User-agent: cohere-training-data-crawler
User-agent: YouBot
User-agent: AI2Bot
User-agent: Diffbot
User-agent: PetalBot
User-agent: Timpibot
User-agent: ImagesiftBot
User-agent: Bytespider
Allow: /
Disallow: /app.php
Disallow: /account.php
Disallow: /decks.php
Disallow: /api/
Allow: /api/tour.php
Allow: /api/demo.php
Disallow: /go/

# The index first, then every file it holds. The index is enough for a crawler
# that reads one; the individual lines are for the several that do not follow a
# sitemapindex from robots.txt, and for pasting into a webmaster console.
#
# sitemap.xml is COMPILED — tools/build_sitemap.py, gated by --check. It was hand
# maintained until 2026-08-18, and an audit that day found 16 live textbook
# chapters and 5 mahjong suit pages listed in no sitemap at all, which also meant
# tools/seo_nightly.sh had never submitted them to IndexNow. Add a page by adding
# its route; the builder finds it.
Sitemap: https://graphchinese.com/sitemap-index.xml
Sitemap: https://graphchinese.com/sitemap.xml
Sitemap: https://graphchinese.com/sitemap-zh.xml
Sitemap: https://graphchinese.com/sitemap-ja.xml
Sitemap: https://graphchinese.com/sitemap-grammar.xml
Sitemap: https://graphchinese.com/sitemap-hsk.xml
Sitemap: https://graphchinese.com/sitemap-dict.xml
Sitemap: https://graphchinese.com/sitemap-readings.xml
Sitemap: https://graphchinese.com/sitemap-shareables.xml

# LLM-friendly summaries of this site, and when to use them
# https://graphchinese.com/llms.txt
# https://graphchinese.com/llms-full.txt
# https://graphchinese.com/AGENTS.md