# ONE group, many user-agents — do not split it.
# Under RFC 9309 (and Google's parser) a crawler obeys ONLY the most specific
# group that names it and ignores every other group. Until 2026-09-17 Googlebot,
# Bingbot, GPTBot, PerplexityBot and ClaudeBot each had their own group holding
# just "Allow: /", which silently discarded the whole Disallow list below for
# exactly the crawlers it was written for (verified live: /admin/, /login and
# /book/* were all fetchable as Googlebot). Naming a bot as an extra User-agent
# line of THIS group keeps it explicitly welcomed while giving it the same rules
# as everyone else. Never add a separate "User-agent: X" + "Allow: /" group.
User-agent: *
User-agent: Googlebot
User-agent: Bingbot
User-agent: Twitterbot
User-agent: facebookexternalhit
User-agent: GPTBot
User-agent: PerplexityBot
User-agent: ClaudeBot
# AI agent resources, explicitly open. "Allow" is a real robots.txt directive
# (unlike a bare "LLMs-txt:" field, which validators reject), so these lines
# both permit and advertise the files.
Allow: /llms.txt
Allow: /llms-full.txt
Allow: /ai.txt
# Content Signals (contentsignals.org / draft-romm-aipref-contentsignals):
# RidePrise WANTS maximum AI visibility — allow search indexing, use as input
# for AI-generated answers (RAG/grounding), and training. This is deliberately
# permissive (the opposite of Cloudflare's old managed ai-train=no).
Content-Signal: search=yes, ai-train=yes, ai-input=yes
Disallow: /admin/
Disallow: /inbox
Disallow: /inbox-sw.js
Disallow: /partner/
Disallow: /book/
Disallow: /it/book/
Disallow: /sq/book/
Disallow: /booking/
Disallow: /it/booking/
Disallow: /sq/booking/
Disallow: /booking-response
Disallow: /leave-review
Disallow: /login
# /my-booking is NOT disallowed, on purpose. GSC reported it "Submitted and
# indexed" (last crawled 2026-08-28), and a Disallow on an already-indexed URL
# is counter-productive: Googlebot stops fetching it, never sees the
# X-Robots-Tag: noindex the server now sends, and the URL can sit in the index
# indefinitely. Crawling is the only way the noindex gets read. The header is
# the real control here; robots.txt is for the paths Google has never seen
# (/admin/, /partner/, /book/*), where it prevents an unbounded crawl trap.
# Re-add this line only once GSC reports the URL as dropped, and not before.
# "Allow: /" LAST on purpose. Google and Bing pick the longest matching rule,
# so order is irrelevant to them — but first-match parsers (Python's
# urllib.robotparser among them) would have let "Allow: /" swallow every
# Disallow above if it came first.
Allow: /

Sitemap: https://rideprise.com/sitemap.xml
# Non-standard discovery hints (kept as comments — bare "LLMs-txt:"/"AI-txt:"
# fields are rejected by robots.txt validators as unknown directives):
# llms.txt: https://rideprise.com/llms.txt
# ai.txt:   https://rideprise.com/ai.txt