# robots.txt for UniMock - University Mock Interview Platform
#
# IMPORTANT — Hostinger HCDN enforcement note (Jun 13 2026):
# Hostinger's CDN edge enforces robots.txt as a firewall: any path matched
# by `Disallow:` returns 401 + body "Restricted by robots.txt" to *real
# users*, not just crawlers. Real-user evidence in LogRocket session
# 6-019ebe58-1871-72f5-b609-44c21bdad5f1 — public endpoints (/api/ping,
# /api/universities, /api/experiences) returned 401 from the edge,
# which made the SPA flash "We'll be back shortly".
#
# So: NEVER write a broad `Disallow: /api/` here. Use specific disallows
# for genuinely private API paths only. Public endpoints (/api/ping,
# /api/auth/*, /api/blog/*, /api/universities, etc.) must remain
# unblocked at this layer.

User-agent: *
Allow: /
Disallow: /dashboard/
Disallow: /admin/
Disallow: /agent/
Disallow: /interviewer/
Disallow: /session/
Disallow: /mock-interview/
Disallow: /result/
Disallow: /api/admin/
Disallow: /api/agent/
Disallow: /api/sessions/
Disallow: /api/applications/
Disallow: /api/users/
Disallow: /api/payments/

# Allow public pages
Allow: /
Allow: /blog
Allow: /blog/*
Allow: /community
Allow: /about
Allow: /for-agents
Allow: /pre-cas-interview
Allow: /cas-interview
Allow: /cas-interview-practice
Allow: /gs-interview
Allow: /sample-pre-cas-questions
Allow: /sample-cas-interview-questions
Allow: /pre-cas-vs-cas-interview
Allow: /what-is-credibility-interview
Allow: /how-ai-mock-interview-works
Allow: /privacy-policy
Allow: /terms-of-service
Allow: /cookie-policy

# Sitemap locations
Sitemap: https://unimock.ai/sitemap.xml
Sitemap: https://unimock.ai/sitemap-images.xml

# GEO: curated guide index for AI crawlers (also at https://unimock.ai/llms.txt)
# LLMs-Txt: https://unimock.ai/llms.txt

# Crawl-delay for respectful crawling
Crawl-delay: 1

# AI Search Engine Bots
# Per-bot rules use the same narrow disallows as User-agent: * — broad
# `Disallow: /api/` would cause Hostinger HCDN to reject real users on
# public API endpoints (see header note above). Crawlers don't follow
# JSON responses anyway, so listing /api/admin/, /api/agent/ etc. is
# sufficient for SEO intent.

User-agent: GPTBot
Allow: /
Disallow: /dashboard/
Disallow: /admin/
Disallow: /agent/
Disallow: /api/admin/
Disallow: /api/agent/
Disallow: /api/sessions/
Disallow: /api/users/

User-agent: ChatGPT-User
Allow: /
Disallow: /dashboard/
Disallow: /admin/
Disallow: /agent/
Disallow: /api/admin/
Disallow: /api/agent/
Disallow: /api/sessions/
Disallow: /api/users/

User-agent: CCBot
Allow: /
Disallow: /dashboard/
Disallow: /admin/
Disallow: /agent/
Disallow: /api/admin/
Disallow: /api/agent/
Disallow: /api/sessions/
Disallow: /api/users/

User-agent: anthropic-ai
Allow: /
Disallow: /dashboard/
Disallow: /admin/
Disallow: /agent/
Disallow: /api/admin/
Disallow: /api/agent/
Disallow: /api/sessions/
Disallow: /api/users/

User-agent: Claude-Web
Allow: /
Disallow: /dashboard/
Disallow: /admin/
Disallow: /agent/
Disallow: /api/admin/
Disallow: /api/agent/
Disallow: /api/sessions/
Disallow: /api/users/

User-agent: PerplexityBot
Allow: /
Disallow: /dashboard/
Disallow: /admin/
Disallow: /agent/
Disallow: /api/admin/
Disallow: /api/agent/
Disallow: /api/sessions/
Disallow: /api/users/

User-agent: Applebot-Extended
Allow: /
Disallow: /dashboard/
Disallow: /admin/
Disallow: /agent/
Disallow: /api/admin/
Disallow: /api/agent/
Disallow: /api/sessions/
Disallow: /api/users/