# Robots.txt for tap.giving
# Policy: maximally permissive. Every public page, including every query-string
# variant of it, is crawlable. Duplicate consolidation is handled by the
# rel=canonical tag on each page, NOT by robots.txt.
#
# WHY (2026-07-28): the old file disallowed /*?*utm_source=* (plus utm_medium,
# utm_campaign, fbclid, gclid). Our own blog soft-CTAs link internally to
# /order.html and /pricing.html WITH utm tags, so Search Console reported our
# two highest-value conversion pages as "Blocked by robots.txt" (19 URLs). A
# blocked URL cannot pass its canonical, so Google could not fold those hits
# back into /order and /pricing. Blocking crawl to prevent duplicates is the
# wrong tool. Canonicals are the right one, and both pages already carry them.
#
# NOTE ON GROUPING (RFC 9309): rules attach to the User-agent group
# directly above them. Disallow rules must live INSIDE the "*" group
# below — placing them after the AI-crawler groups silently scopes
# them to the last-listed bot only.
#
# Before adding ANY Disallow here, confirm nothing on the site links to the
# pattern. If we link to it, we want it crawled.

User-agent: *
Allow: /
# The only blocks: non-public surfaces that are not pages at all.
Disallow: /admin/
Disallow: /private/
Disallow: /_template.html
# Worker API endpoints. POST-only, so a crawler GET returns 405 and Search
# Console files it under "Blocked due to other 4xx". Not a page, never linked
# as an href, nothing to index. This is the one block that stays.
Disallow: /api/

# Sitemap location
Sitemap: https://tap.giving/sitemap.xml

# AI Agent Discovery
# See https://llmstxt.org for specification
# llms.txt provides structured context for AI models (ChatGPT, Claude, Gemini, Perplexity)
# This helps AI assistants accurately answer questions about Tap.Giving, NFC giving,
# tap-to-give plates, and church giving technology

# Explicitly welcome AI crawlers (full access, no restrictions)
User-agent: GPTBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: anthropic-ai
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: Claude-Web
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Google-Extended
Allow: /

User-agent: Amazonbot
Allow: /

User-agent: cohere-ai
Allow: /