# =============================================================================
# robots.txt — Kawartha Corners Food Corral
#
# Allows major search engines (the site earns real traffic from Google), and
# blocks AI-training crawlers and aggressive SEO scrapers.
#
# Note: robots.txt is advisory only — it is NOT a security control. Well-behaved
# bots obey it; malicious scanners ignore it entirely. The WordPress/PHP probe
# traffic hitting this site is handled by the 410 redirect rules in netlify.toml.
# =============================================================================

# ── Major search engines ─────────────────────────────────────────────────────
User-agent: Googlebot
Allow: /

User-agent: Bingbot
Allow: /

User-agent: Slurp
Allow: /

User-agent: DuckDuckBot
Allow: /

# ── Block AI training / scraper bots ─────────────────────────────────────────
User-agent: GPTBot
Disallow: /

User-agent: ChatGPT-User
Disallow: /

User-agent: OAI-SearchBot
Disallow: /

User-agent: CCBot
Disallow: /

User-agent: anthropic-ai
Disallow: /

User-agent: Claude-Web
Disallow: /

User-agent: ClaudeBot
Disallow: /

User-agent: Google-Extended
Disallow: /

User-agent: Applebot-Extended
Disallow: /

User-agent: Bytespider
Disallow: /

User-agent: PetalBot
Disallow: /

User-agent: Amazonbot
Disallow: /

User-agent: meta-externalagent
Disallow: /

# ── Block aggressive SEO / scraper bots ──────────────────────────────────────
User-agent: AhrefsBot
Disallow: /

User-agent: SemrushBot
Disallow: /

User-agent: MJ12bot
Disallow: /

User-agent: DotBot
Disallow: /

User-agent: BLEXBot
Disallow: /

User-agent: MegaIndex.ru
Disallow: /

User-agent: ia_archiver
Disallow: /

# ── Default: allow everything else ───────────────────────────────────────────
# The disallowed paths below are WordPress/PHP endpoints that do not exist on
# this React site. Listing them tells well-behaved crawlers not to probe them.
User-agent: *
Disallow: /wp-admin
Disallow: /wp-login.php
Disallow: /wp-includes/
Disallow: /wp-content/
Disallow: /wp-json/
Disallow: /wp-cron.php
Disallow: /wp-config.php
Disallow: /setup-config.php
Disallow: /xmlrpc.php
Disallow: /wlwmanifest.xml
Disallow: /wordpress
Disallow: /index.php
Disallow: /.env
Disallow: /.git/
Disallow: /phpmyadmin
Disallow: /pma
Disallow: /adminer
Disallow: /cgi-bin/
Disallow: /phpinfo.php
Disallow: /administrator
Allow: /

# ── Sitemap ──────────────────────────────────────────────────────────────────
Sitemap: https://kawarthacornersfoodcorral.ca/sitemap.xml