# PDFops — robots.txt
#
# Posture: open to citation-class AI crawlers (search-index + on-demand
# user fetch) and training-class crawlers; blocked to Bytespider.
# Reference: docs/decisions/0002-initial-geo-improvements.md.

User-agent: *
Allow: /
# /api/ endpoints are not content — and GET /api/checkout creates a real
# Stripe checkout session, so crawlers following the pricing-page Buy
# links were minting sessions that expired 24h later (alert noise,
# 2026-08-06). Docs pages describe every endpoint; nothing is lost.
Disallow: /api/

# --- Citation-class crawlers (search index + on-demand user fetch) ---
# These power AI-search results and answer-citation. Explicit Allow is
# not strictly required (the wildcard above covers them), but several
# of these honor explicit directives more reliably than wildcard
# fallthrough, and the listing is itself a discoverable signal of
# editorial intent.

User-agent: OAI-SearchBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: Claude-SearchBot
Allow: /

User-agent: Claude-Web
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: Perplexity-User
Allow: /

User-agent: Applebot
Allow: /

User-agent: GoogleOther
Allow: /

User-agent: Google-NotebookLM
Allow: /

User-agent: Meta-ExternalAgent
Allow: /

User-agent: DuckAssistBot
Allow: /

User-agent: MistralAI-User
Allow: /

User-agent: Amazonbot
Allow: /

# --- Training-class crawlers (corpus collection) ---
# Allowed during the discovery window; paid tiers are live — kept allowed (AI answer engines are the ICP).

User-agent: GPTBot
Allow: /

User-agent: ClaudeBot
Allow: /

User-agent: anthropic-ai
Allow: /

User-agent: Google-Extended
Allow: /

User-agent: Applebot-Extended
Allow: /

User-agent: CCBot
Allow: /

User-agent: cohere-training-data-crawler
Allow: /

# --- Blocked: aggressive, no citation surface ---
# Bytespider is known to ignore robots.txt; the Disallow is a polite
# signal, real enforcement requires an edge-WAF rule (deferred — Vercel
# Firewall custom rules are paid-tier; revisit if Bytespider traffic
# shows up in logs).

User-agent: Bytespider
Disallow: /

Sitemap: https://pdfops.dev/sitemap.xml
Host: pdfops.dev

# --- IndexNow ---
# Key file: /203683cec395429b8e8acfaacaa20c8d.txt
IndexNow: 203683cec395429b8e8acfaacaa20c8d