# ------------------------------
# SEARCH CRAWLERS
# Named agents below are explicitly allowed. Googlebot belongs to this role
# and follows the general site rules below.
# ------------------------------
User-agent: OAI-SearchBot
Allow: /

# Bing index powers Bing Search, Microsoft Copilot answers, and ChatGPT search retrieval
User-agent: bingbot
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: PerplexityBot
Allow: /

# ------------------------------
# SEARCH PREVIEW FETCHERS (explicit allow)
# ------------------------------
User-agent: BingPreview
Allow: /

User-agent: MicrosoftPreview
Allow: /

# ------------------------------
# USER-DIRECTED FETCHERS (explicit allow)
# These fetch at a user's direction rather than crawling automatically.
# Provider behavior around robots.txt differs for user-directed requests.
# ------------------------------
User-agent: ChatGPT-User
Allow: /

User-agent: Claude-User
Allow: /

User-agent: Perplexity-User
Allow: /

# ------------------------------
# TRAINING OR BROADER MODEL USE (TempGuru policy: allow)
# Access is a business-policy decision separate from search visibility.
# Google-Extended is a robots control token, not a standalone HTTP crawler.
# It does not affect Google Search rankings or AI Overviews / AI Mode eligibility.
# ------------------------------
User-agent: GPTBot
Allow: /

User-agent: ClaudeBot
Allow: /
Crawl-delay: 10

User-agent: Google-Extended
Allow: /

# ------------------------------
# OTHER AUTOMATED DISCOVERY AND FETCH AGENTS (explicit allow)
# ------------------------------
User-agent: BraveBot
Allow: /

User-agent: anthropic-ai
Allow: /

User-agent: Kimi-SearchBot
Allow: /

User-agent: Kimi-User
Allow: /

User-agent: MistralAI-Index
Allow: /

User-agent: MistralAI-User
Allow: /

User-agent: Qwantbot
Allow: /

User-agent: MojeekBot
Allow: /

User-agent: SeznamBot
Allow: /

User-agent: YandexBot
Allow: /

# Bytespider (ByteDance/TikTok)
User-agent: Bytespider
Allow: /

# ------------------------------
# CHINESE SEARCH ECOSYSTEM (explicit allow)
# ------------------------------
# Baiduspider - powers Baidu Search + ERNIE Bot (Baidu's LLM)
User-agent: Baiduspider
Allow: /
# YisouSpider - Shenma/UC/Quark (Alibaba ecosystem) - relevant for Qwen / DashScope discoverability
User-agent: YisouSpider
Allow: /
# Sogou web spider - Tencent ecosystem (powers Yuanbao)
User-agent: Sogou web spider
Allow: /
# 360Spider - Qihoo 360 Search
User-agent: 360Spider
Allow: /

# ------------------------------
# OTHER BROADER-USE AND DATASET AGENTS (TempGuru policy: allow)
# ------------------------------
User-agent: GoogleOther
Allow: /

User-agent: Applebot-Extended
Allow: /

User-agent: Amazonbot
Allow: /

User-agent: FacebookBot
Allow: /

User-agent: meta-externalagent
Allow: /

User-agent: meta-externalfetcher
Allow: /

# CommonCrawl dataset access is maintained as a separate business-policy allow.
User-agent: CCBot
Allow: /

# Inert alias (real UA is CCBot/2.0) - kept to document intent.
User-agent: CommonCrawl
Allow: /

# ------------------------------
# GENERAL SITE CRAWL RULES (legacy Squarespace paths retained)
# ------------------------------
User-agent: *
Content-Signal: search=yes, ai-input=yes, ai-train=yes
Disallow: /config
Disallow: /search
Disallow: /account$
Disallow: /account/
Disallow: /commerce/digital-download/
Disallow: /api/
Allow: /api/ui-extensions/
Disallow: /static/
Disallow: /*?format=json
Disallow: /*&format=json
Disallow: /*?format=page-context
Disallow: /*&format=page-context
Disallow: /*?format=main-content
Disallow: /*&format=main-content
Disallow: /*?format=json-pretty
Disallow: /*&format=json-pretty
Disallow: /*?format=ical
Disallow: /*&format=ical
Disallow: /*?reversePaginate=*
Disallow: /*&reversePaginate=*

# ------------------------------
Sitemap: https://tempguru.co/sitemap.xml