# Apply4Me — production robots (canonical host: https://www.apply4me.io)
# Marketing + blog are crawlable; app/auth surfaces are not.
#
# Every app route lives under a locale prefix (/pl/settings, /en/cv-kit, ...),
# so the bare `Disallow: /settings` this file used to carry matched NOTHING a
# crawler actually sees, and every signed-in surface shipped as a crawlable
# 70-word SPA shell. Each segment below is therefore listed once unprefixed
# (legacy URLs + redirect targets) and once per locale.
#
# Deliberately spelled out instead of `/*/settings`: robots patterns match as a
# PREFIX, so the wildcard form would also block a future blog post slugged
# "settings-guide". Locales come from src/i18n/locales.ts (en, pl, de) — adding
# a locale or an app route segment means adding lines here, or it leaks.
#
# NOT blocked here on purpose:
#   - ?page= / ?q= blog variants — they carry a canonical to the real path, and
#     robots-blocking them would stop Google reading that canonical.
#   - /de — noindex is served in the shell HTML instead, which is definitive;
#     a robots block would let the bare URL get indexed with no content.
User-agent: *
Allow: /

# Infra
Disallow: /admin
Disallow: /api
Disallow: /monitoring

# Auth + signup funnel
Disallow: /login
Disallow: /en/login
Disallow: /pl/login
Disallow: /de/login
Disallow: /sign-in
Disallow: /en/sign-in
Disallow: /pl/sign-in
Disallow: /de/sign-in
Disallow: /sign-up
Disallow: /en/sign-up
Disallow: /pl/sign-up
Disallow: /de/sign-up
Disallow: /signup
Disallow: /en/signup
Disallow: /pl/signup
Disallow: /de/signup
Disallow: /auth/
Disallow: /en/auth/
Disallow: /pl/auth/
Disallow: /de/auth/
Disallow: /verify
Disallow: /en/verify
Disallow: /pl/verify
Disallow: /de/verify
Disallow: /plan-selection
Disallow: /en/plan-selection
Disallow: /pl/plan-selection
Disallow: /de/plan-selection
Disallow: /onboarding
Disallow: /en/onboarding
Disallow: /pl/onboarding
Disallow: /de/onboarding
Disallow: /winback
Disallow: /en/winback
Disallow: /pl/winback
Disallow: /de/winback

# Signed-in product surfaces
Disallow: /dashboard
Disallow: /en/dashboard
Disallow: /pl/dashboard
Disallow: /de/dashboard
Disallow: /applications
Disallow: /en/applications
Disallow: /pl/applications
Disallow: /de/applications
Disallow: /cv-kit
Disallow: /en/cv-kit
Disallow: /pl/cv-kit
Disallow: /de/cv-kit
Disallow: /job-recommendations
Disallow: /en/job-recommendations
Disallow: /pl/job-recommendations
Disallow: /de/job-recommendations
Disallow: /auto-apply
Disallow: /en/auto-apply
Disallow: /pl/auto-apply
Disallow: /de/auto-apply
Disallow: /career-path
Disallow: /en/career-path
Disallow: /pl/career-path
Disallow: /de/career-path
Disallow: /career-paths
Disallow: /en/career-paths
Disallow: /pl/career-paths
Disallow: /de/career-paths
Disallow: /interview-prep
Disallow: /en/interview-prep
Disallow: /pl/interview-prep
Disallow: /de/interview-prep
Disallow: /settings
Disallow: /en/settings
Disallow: /pl/settings
Disallow: /de/settings
Disallow: /purchase
Disallow: /en/purchase
Disallow: /pl/purchase
Disallow: /de/purchase

# Utility / internal pages
Disallow: /delete-account
Disallow: /en/delete-account
Disallow: /pl/delete-account
Disallow: /de/delete-account
Disallow: /hero-lab
Disallow: /en/hero-lab
Disallow: /pl/hero-lab
Disallow: /de/hero-lab

# AI answer engines — explicitly allowed.
#
# They already match `User-agent: *` above, so these groups change nothing
# today. They exist because a robots.txt group is winner-takes-all: the moment
# any future edit adds a broad Disallow to `*`, every one of these crawlers
# would inherit it silently and the site would drop out of AI answers with no
# error anywhere. Naming them means that edit has to be deliberate.
User-agent: GPTBot
User-agent: OAI-SearchBot
User-agent: ChatGPT-User
User-agent: ClaudeBot
User-agent: Claude-User
User-agent: Claude-SearchBot
User-agent: PerplexityBot
User-agent: Perplexity-User
User-agent: Google-Extended
User-agent: Applebot-Extended
User-agent: meta-externalagent
User-agent: Bingbot
Allow: /

# Google Ads landing page crawler.
#
# AdsBot-Google ignores the global `User-agent: *` group BY DESIGN — it only
# obeys a group that names it. Without this block it is technically unbound
# here; spelled out so an ad pointed at any URL is checked against a rule we
# chose rather than none. Ads only run to /pl marketing + tool pages.
User-agent: AdsBot-Google
User-agent: AdsBot-Google-Mobile
Allow: /

Sitemap: https://www.apply4me.io/sitemap.xml
Sitemap: https://www.apply4me.io/sitemap-images.xml