# LLM and AI-assistant crawlers are explicitly welcome. Listing them by
# name (instead of relying only on the User-agent: * block below) makes the
# policy unambiguous to operators reviewing this file. Each block repeats
# the same Disallow: /? rule as the wildcard so AI crawlers also skip the
# query-string namespace on the home page — see the longer note further down
# for why that matters. New AI crawlers can be added here as they appear.

User-agent: GPTBot
Allow: /
Disallow: /?

User-agent: ChatGPT-User
Allow: /
Disallow: /?

User-agent: OAI-SearchBot
Allow: /
Disallow: /?

User-agent: ClaudeBot
Allow: /
Disallow: /?

User-agent: Claude-Web
Allow: /
Disallow: /?

User-agent: anthropic-ai
Allow: /
Disallow: /?

User-agent: PerplexityBot
Allow: /
Disallow: /?

User-agent: Perplexity-User
Allow: /
Disallow: /?

User-agent: Google-Extended
Allow: /
Disallow: /?

User-agent: Applebot-Extended
Allow: /
Disallow: /?

User-agent: CCBot
Allow: /
Disallow: /?

User-agent: Bytespider
Allow: /
Disallow: /?

User-agent: Meta-ExternalAgent
Allow: /
Disallow: /?

User-agent: *
Allow: /

# Block crawling of home-page filter/state URLs.
#
# When users interact with the comparison tool we mirror their selections
# (city, banks, restaurants, owned cards, wallet customizations, etc.) into
# the URL as query strings so they can share a link. Every one of these
# URLs already declares <link rel="canonical" href="https://konsacard.pk/">
# so Google does not index them. But Google was still crawling them, eating
# crawl budget that should go to the 1,400+ unique bank/restaurant/card
# pages. Blocking the entire query-string namespace on `/` stops the waste.
#
# This affects only the home page. Bank pages at /banks/<slug>/ and
# restaurant pages at /restaurants/<slug>/ remain fully crawlable.
Disallow: /?

# Be respectful of crawl rate.
Crawl-delay: 1

Sitemap: https://konsacard.pk/sitemap.xml