# Garcia-Windsor, P.C. — robots.txt
# Bilingual (English + Español) Dallas family & business law firm.
# Full crawler access is intentionally permitted for search discovery + AI
# retrieval. Only private admin surfaces and design assets are excluded.

User-agent: *
Allow: /
Disallow: /admin/
Disallow: /api/admin/
Disallow: /styleguide
Disallow: /es/styleguide

# ------------------------------------------------------------
# Explicit allowances for major AI / LLM crawlers.
# Listing each one avoids ambiguity vs. a generic User-agent: *,
# and lets us change grants per bot if licensing terms shift.
# ------------------------------------------------------------

# OpenAI: general-purpose training + inference crawler.
User-agent: GPTBot
Allow: /
Disallow: /admin/
Disallow: /api/admin/
Disallow: /styleguide

# OpenAI: on-demand fetch when a ChatGPT user asks about a URL.
User-agent: ChatGPT-User
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# OpenAI: SearchGPT indexing crawler.
User-agent: OAI-SearchBot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# Anthropic (Claude) crawlers.
User-agent: ClaudeBot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

User-agent: Claude-Web
Allow: /
Disallow: /admin/
Disallow: /api/admin/

User-agent: anthropic-ai
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# Google's separate opt-in signal for Gemini / AI Overviews training.
User-agent: Google-Extended
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# Perplexity retrieval + training.
User-agent: PerplexityBot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

User-agent: Perplexity-User
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# Common Crawl (feeds many downstream LLMs).
User-agent: CCBot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# Apple Intelligence.
User-agent: Applebot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

User-agent: Applebot-Extended
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# Amazon Alexa / product AI.
User-agent: Amazonbot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# Meta AI crawler.
User-agent: Meta-ExternalAgent
Allow: /
Disallow: /admin/
Disallow: /api/admin/

User-agent: Meta-ExternalFetcher
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# You.com search + assistant.
User-agent: YouBot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# Diffbot, Cohere, other AI retrieval crawlers.
User-agent: Diffbot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

User-agent: cohere-ai
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# ------------------------------------------------------------
# Traditional search engine crawlers (redundant with * but explicit).
# ------------------------------------------------------------

User-agent: Googlebot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

User-agent: Bingbot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

User-agent: DuckDuckBot
Allow: /
Disallow: /admin/
Disallow: /api/admin/

# ------------------------------------------------------------
# Sitemap
# The canonical production URL is used so the same robots.txt file works on
# preview + production without swapping. Google follows Sitemap: directives
# regardless of same-origin.
#
# Layout:
#   * sitemap.xml     — top-level index (registered in Search Console + Bing)
#   * sitemap-en.xml  — every English canonical URL with hreflang alternates
#   * sitemap-es.xml  — every Spanish canonical URL with hreflang alternates
# Listing all three gives search engines an explicit fallback if the index is
# temporarily unreachable.
# ------------------------------------------------------------

Sitemap: https://www.garciawindsor.com/sitemap.xml
Sitemap: https://www.garciawindsor.com/sitemap-en.xml
Sitemap: https://www.garciawindsor.com/sitemap-es.xml

# ------------------------------------------------------------
# LLM content manifests (Answer.AI llms.txt proposed standard)
# ------------------------------------------------------------
# Curated Markdown indexes of the site's most authoritative content, aimed
# at AI answer engines (ChatGPT Search, Perplexity, Google AI Overviews,
# Claude). See https://llmstxt.org for the spec.
#
#   * llms.txt       — compact bilingual index with descriptions.
#   * llms-full.txt  — deep-context concatenation of every service page
#                      and blog post (Markdown, single retrieval pass).
# Both files are auto-regenerated on backend boot from the site's content
# source-of-truth modules, so they cannot drift from the live site.

# LLM-Content: https://www.garciawindsor.com/llms.txt
# LLM-Full-Content: https://www.garciawindsor.com/llms-full.txt