# RSA Development — robots policy
# Posture: fully open to search + AI answer engines. We *want* to be cited.
#
# IMPORTANT (REP semantics): a crawler obeys ONE group — the most specific
# matching User-agent — and does NOT inherit rules from the `*` group. Every
# named crawler is therefore listed in the SAME group as the shared rules
# below, so the private-path exclusions apply to all of them. Adding a
# separate `User-agent: Googlebot / Allow: /` group would have silently
# opened /admin, /api and /email to Google, Bing and every AI crawler.
#
# robots.txt is neither access control nor a reliable noindex: the admin area
# stays behind authentication and private routes keep their server-side
# noindex. CSS, JS and images are intentionally never blocked, and no
# redirect target is blocked, so Google can render pages fully.

User-agent: *
# ---- Major search crawlers ----
User-agent: Googlebot
User-agent: Googlebot-Image
User-agent: Googlebot-News
User-agent: GoogleOther
User-agent: Google-Extended
User-agent: Bingbot
User-agent: DuckDuckBot
User-agent: DuckAssistBot
User-agent: Applebot
User-agent: Applebot-Extended
User-agent: YandexBot
User-agent: Baiduspider
User-agent: PetalBot
# ---- AI answer engines & training crawlers (explicit opt-in) ----
# We want RSA Development surfaced inside ChatGPT, Claude, Perplexity,
# Gemini, Copilot, Apple Intelligence, Meta AI and Amazon Rufus answers
# about California ADUs.
User-agent: GPTBot
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: ClaudeBot
User-agent: Claude-Web
User-agent: Claude-SearchBot
User-agent: anthropic-ai
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Meta-ExternalAgent
User-agent: Meta-ExternalFetcher
User-agent: FacebookBot
User-agent: facebookexternalhit
User-agent: Amazonbot
User-agent: YouBot
User-agent: cohere-ai
User-agent: cohere-training-data-crawler
User-agent: Bytespider
User-agent: TikTokSpider
User-agent: Diffbot
User-agent: CCBot
User-agent: MistralAI-User
User-agent: Kagibot
User-agent: Timpibot
# Exclusions first: Google applies longest-match, but older first-match
# parsers (e.g. Python's urllib RobotFileParser) would otherwise let
# `Allow: /` win over these rules. This order is correct under both.
Disallow: /admin
Disallow: /admin/
Disallow: /lovable/
Disallow: /api/
Disallow: /unsubscribe
Disallow: /email/
Allow: /

# ---- Sitemaps (search + AI discovery) ----
Sitemap: https://rsa-development.com/sitemap.xml
Sitemap: https://rsa-development.com/sitemap-core.xml
Sitemap: https://rsa-development.com/sitemap-images.xml

# ---- AI discovery hints (not part of REP but widely read) ----
# Machine-readable site summary for LLMs:  https://rsa-development.com/llms.txt
# Extended training/answer corpus:         https://rsa-development.com/llms-full.txt
# Content usage policy for AI:             https://rsa-development.com/ai.txt
Sitemap: https://rsa-development.com/sitemap-cities.xml