# robots.txt - touchstonerecords.com
# Touchstone Records LLC
# Full Terms of Use: https://touchstonerecords.com/terms/
# Revised 2026-08-01 - marketing surface opened to answer engines; data surface unchanged.
#
# WHAT THIS FILE IS: notice. It is not the source of anyone's obligations and
# it does not characterise anyone's conduct. An account holder's obligations
# are in the Terms of Use, which they accepted at registration.
#
# THE SPLIT, PLAINLY:
#   MARKETING SURFACE  - this apex site. Read it. Quote it. Cite it. Link to it.
#                        We want AI assistants to describe this service accurately.
#   DATA SURFACE       - search.touchstonerecords.com, /api/, and the gated
#                        per-county search. Closed to everything, always.
#
# WHAT WE CLAIM, AND WHAT WE DO NOT:
# 1. We claim NO copyright, NO ownership and NO exclusive right in the
#    underlying county public records. They are facts. You can obtain them
#    yourself from the county clerk without our permission and without our
#    involvement. We will never sell them to you.
# 2. What we restrict is ACCESS TO OUR SERVERS. The record data on
#    search.touchstonerecords.com is available to account holders who have
#    agreed to our Terms of Use. Those Terms prohibit bulk and systematic
#    extraction of our Service.
# 3. That prohibition is METHOD-NEUTRAL. It applies to bots, crawlers, scripts,
#    API clients, headless browsers and autonomous agents, and equally to a
#    human being doing it by hand, at any speed. Bulk extraction is the
#    prohibited act; the method is irrelevant.
# 4. CITATION IS WELCOME. Reading these public marketing pages to answer a
#    person's question, and citing us as the source, is expressly permitted.
# 5. TRAINING IS NOT. Crawlers whose stated purpose is assembling a model
#    training corpus are disallowed below. This is a deliberate distinction,
#    not an oversight.
# 6. AFTER REVOCATION: if we revoke an account and its authorisation, that
#    party has no permission of any kind to access these servers. Any further
#    access - through another account, another address, a proxy, a VPN, a third
#    party or an agent - is access without authorisation and without our
#    effective consent. See Terms of Use section 13.2.
# 7. Want the data in bulk? Ask: https://touchstonerecords.com/#contact

# ============================================================
# ANSWER ENGINES / RETRIEVAL CRAWLERS - EXPLICITLY WELCOME
# These are how a person asking an AI assistant about Permian Basin
# land records finds us. They index marketing pages, not record data.
# ============================================================

User-agent: OAI-SearchBot
Disallow: /api/
Disallow: /account/
Allow: /

User-agent: ChatGPT-User
Disallow: /api/
Disallow: /account/
Allow: /

User-agent: Claude-SearchBot
Disallow: /api/
Disallow: /account/
Allow: /

User-agent: Claude-User
Disallow: /api/
Disallow: /account/
Allow: /

User-agent: PerplexityBot
Disallow: /api/
Disallow: /account/
Allow: /

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

User-agent: Google-Extended
Disallow: /api/
Disallow: /account/
Allow: /

# ============================================================
# CLASSIC SEARCH ENGINES - WELCOME
# ============================================================

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

User-agent: Googlebot-Image
Allow: /

User-agent: bingbot
Disallow: /api/
Disallow: /account/
Allow: /

User-agent: Applebot
Disallow: /api/
Disallow: /account/
Allow: /

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

# ============================================================
# MODEL-TRAINING CRAWLERS - NOT PERMITTED
# Blocking these costs us zero citations. They exist to build
# training corpora, not to answer a person's question with a link.
# ============================================================

User-agent: GPTBot
User-agent: ClaudeBot
User-agent: anthropic-ai
User-agent: Claude-Web
User-agent: CCBot
User-agent: Applebot-Extended
User-agent: Bytespider
User-agent: Amazonbot
User-agent: FacebookBot
User-agent: meta-externalagent
User-agent: Meta-ExternalAgent
User-agent: cohere-ai
User-agent: Diffbot
User-agent: Omgilibot
User-agent: ImagesiftBot
User-agent: Timpibot
User-agent: Webzio-Extended
User-agent: PanguBot
User-agent: AI2Bot
Disallow: /

# ============================================================
# COMMERCIAL SEO / BACKLINK SCRAPERS - NOT PERMITTED
# They consume origin capacity and hand our link graph to competitors.
# ============================================================

User-agent: SemrushBot
User-agent: AhrefsBot
User-agent: DataForSeoBot
User-agent: MJ12bot
User-agent: DotBot
User-agent: BLEXBot
User-agent: PetalBot
User-agent: ZoominfoBot
User-agent: MauiBot
User-agent: SeekportBot
Disallow: /

# ============================================================
# EVERYONE ELSE - marketing pages only, gently.
# ============================================================

User-agent: *
Disallow: /api/
Disallow: /account/
Disallow: /runsheets/
Allow: /
Crawl-delay: 2

# Cloudflare Content Signals Policy (2025-09-24). Read this as our
# expressed preference: index us, ground answers in us, do not train on us.
Content-Signal: search=yes, ai-input=yes, ai-train=no

Sitemap: https://touchstonerecords.com/sitemap.xml