User-agent: *
Allow: /
Allow: /pricing
Allow: /articles
Allow: /articles/
# Research feed — aggregators and readers must be able to reach it, and the
# catch-all Disallow rules below are prefix matches.
Allow: /feed.xml
Allow: /rss.xml
Allow: /feed
Allow: /llms.txt
Allow: /llms-full.txt
# Keep static assets crawlable — the favicon is what Google uses for the
# search-result icon, and the catch-all `Disallow: /*?*` below would otherwise
# block any versioned (?v=) asset (favicon, CSS, JS) from Googlebot.
Allow: /static/
# SHARE URLS CARRY QUERY STRINGS (found 2026-08-13, the X card that never
# rendered): every link the rollout distributes is utm-tagged, and the
# catch-all `Disallow: /*?*` below blocked Twitterbot from crawling exactly
# those URLs — so a shared paper could not produce its card. Longest-match
# wins under Google-style robots matching, so these carve the public reading
# surfaces (with or without query) out of the catch-all while everything the
# catch-all actually protects stays blocked.
Allow: /research
Allow: /research/
Allow: /research/*?*
Allow: /articles/*?*
# `Allow: /?*` USED TO SIT HERE AND DID NOTHING (2026-09-08). Google picks the
# LONGEST matching rule: /?* is three characters, `Disallow: /*?*` below is four,
# so the catch-all won and Search Console reported /?ref=producthunt blocked with
# /*?* named as the rule. Campaign and share links land on the ROOT, and the root
# with a query serves the homepage carrying a canonical to https://quanterlab.com/,
# so crawling it is what consolidates the link. These forms match the root ONLY and
# are long enough to win. Do NOT write them as /*?ref= — that matches every path and
# would out-rank `Disallow: /api/`.
Allow: /?ref=*
Allow: /?utm_*
Allow: /?r=*
Allow: /?fbclid=*
Allow: /?gclid=*
Disallow: /desktop/releases/
Disallow: /desktop/updates/
Disallow: /home
Disallow: /api/
Disallow: /login
Disallow: /register
Disallow: /forgot-password
Disallow: /reset-password/
Disallow: /verify-email/
Disallow: /verification-pending
Disallow: /resend-verification
Disallow: /logout
Disallow: /strategy/
Disallow: /fundamental/
Disallow: /paper-trading/
Disallow: /ml-strategy/
Disallow: /admin/
Disallow: /mcp
Disallow: /oauth/
Disallow: /primitives
Disallow: /primitives/
Disallow: /lab
Disallow: /*?*

# AI crawlers — explicitly allow so QuanterLab is visible to LLMs
# (ChatGPT search, Claude web tool, Perplexity, Google AI Overviews).
# Default `User-agent: *` already covers them, but several boilerplates
# ship with implicit denies, so make the intent explicit.
User-agent: GPTBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: OAI-SearchBot
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: Claude-Web
Allow: /

User-agent: anthropic-ai
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Google-Extended
Allow: /

Sitemap: https://quanterlab.com/sitemap.xml