# Everything is open to everyone, AI included. That is deliberate: this site
# exists to be read and cited.
#
# Two honest notes about what follows.
#
# 1. Absent a Disallow, every crawler is already permitted. So the named groups
#    below are documentation of intent, not a functional change — they do not
#    unlock anything the wildcard group had not already allowed.
#
# 2. If you ever restrict something: under RFC 9309 a crawler obeys only the
#    MOST SPECIFIC group matching its token and ignores all others, including
#    `*`. A Disallow added to `*` will NOT apply to any bot named below. You
#    would have to add it to each group as well. Unrecognised tokens are inert,
#    so a stale name here is harmless.

User-agent: *
Allow: /

# Affirmative usage preferences, per contentsignals.org. Unlike the Allow rules
# above this actually says something new: not merely "you may fetch this" but
# "you may train on it, ground answers in it, and index it".
Content-Signal: search=yes, ai-input=yes, ai-train=yes

# -- Model training ----------------------------------------------------------

User-agent: GPTBot
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: Google-Extended
Allow: /

User-agent: Applebot-Extended
Allow: /

User-agent: MistralAI-Training
Allow: /

User-agent: meta-externalagent
Allow: /

User-agent: Amazonbot
Allow: /

User-agent: Bytespider
Allow: /

User-agent: CCBot
Allow: /

User-agent: AI2Bot
Allow: /

User-agent: cohere-training-data-crawler
Allow: /

User-agent: Webzio-Extended
Allow: /

User-agent: omgili
Allow: /

# -- Retrieval and answer engines --------------------------------------------
# These fetch a page to answer a question now. They are the ones that actually
# produce citations back here, so they matter most.

User-agent: OAI-SearchBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: OAI-AdsBot
Allow: /

User-agent: Claude-User
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Perplexity-User
Allow: /

User-agent: MistralAI-User
Allow: /

User-agent: MistralAI-Index
Allow: /

User-agent: DuckAssistBot
Allow: /

User-agent: Amzn-SearchBot
Allow: /

User-agent: Amzn-User
Allow: /

User-agent: Applebot
Allow: /

User-agent: YouBot
Allow: /

User-agent: PetalBot
Allow: /

User-agent: Diffbot
Allow: /

User-agent: Diffbot-User
Allow: /

# Deliberately absent:
#   xAI / Grok      — no documented robots.txt token exists. The strings doing
#                     the rounds (xAI-Bot, GrokBot, xAI-Grok) trace only to SEO
#                     content farms that contradict each other, and x.ai's own
#                     robots.txt defines no xAI token. Inventing one is noise.
#   Copilot         — Microsoft states there is no Copilot token; Bing Chat uses
#                     bingbot's index, and the control surface is the NOARCHIVE
#                     and NOCACHE meta tags. This site sets neither, which is
#                     what allows Copilot to use and cite it.
#   Anything with a space in the token (e.g. "Sogou web spider") — violates the
#                     RFC 9309 product-token grammar and may silently not match.

Sitemap: https://jays.fyi/sitemap-index.xml

# Markdown index of the site, per the llmstxt.org proposal. Every page also has
# a markdown version at its own URL plus `.md`, which is the cheaper thing to
# read: https://jays.fyi/llms.txt