# AI crawlers (OAI-SearchBot, ChatGPT-User, GPTBot, Claude-SearchBot, PerplexityBot,
# Bingbot, ...) are deliberately covered by the blanket allow below: ChatGPT/AI-assistant
# recommendations are VidWords' largest traffic source, and OpenAI documents that sites
# opting out of OAI-SearchBot are not shown in ChatGPT search. Allowing training crawlers
# (GPTBot, CCBot, ...) is an explicit owner decision, not an oversight — in-model brand
# familiarity is worth more to this business than a training opt-out.
User-agent: *
Allow: /
Disallow: /app/
Disallow: /api/
Disallow: /admin
Disallow: /verify
# Transcript library is OFF (PUBLIC_TRANSCRIPT_PAGES=0, set 2026-07-21): /watch/:id and
# /transcripts now 404 and are absent from the sitemap index, so no Disallow is needed —
# a 404 removes them from an index; a robots block would only stop the recrawl that
# confirms they are gone. Turning it back on is a one-line env change.
# Why off: across 13,909 tracked pageviews the library drew 1, and GSC recorded 0 clicks
# from 26 impressions — while it was the site's whole exposure to Google's scaled-content
# policy (republished third-party captions) and to DMCA risk.
# Only user-generated Watch share pages need an explicit block.
Disallow: /watch/shared/

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

# There is no registered directive for advertising llms.txt, so this comment is the
# only discovery hint short of guessing the path. Both files serve 200 text/plain.
# llms.txt: https://vidwords.com/llms.txt
# llms-full.txt: https://vidwords.com/llms-full.txt