User-agent: *
Allow: /

Sitemap: https://tts.ai/sitemap.xml

# Disallow admin and API endpoints
Disallow: /admin/
Disallow: /api/accounts/
Disallow: /api/verify/
Disallow: /api/session-token/
Disallow: /ipns/

# Tracking + UI-state query params. Variants the canonical strip-list
# and CanonicalRedirectMiddleware also handle, but blocking crawl
# removes the wasted budget entirely and stops Google indexing them
# as duplicates of the bare path.
#
# `?ref=`     user-referral / signup attribution.
# `?return=`  language picker return-path.
# `?lang=en`  redundant (en is the default).
# `?model=`,  `?voice=`, `?agent=`: content-pre-select query params that
#             render the SAME page body with a filter applied.
#             /voices/?model=<slug> was a SpamBrain feeder for 44 URLs
#             until 2026-05-14 (removed from sitemap; blocked here so
#             pre-selected deep-links into the picker don't get indexed
#             as duplicates either).
# `?language=`, `?gender=` — filter facets on /voices/.
# `?sort=`, `?new=`, `?is_free=`, `?tool=`, `?from=`, `?next=` — generic
#             UI-state from across the fleet's playbook list.
Disallow: /*?ref=
Disallow: /*&ref=
Disallow: /*?return=
Disallow: /*&return=
Disallow: /*?lang=en
Disallow: /*&lang=en
Disallow: /*?model=
Disallow: /*&model=
Disallow: /*?voice=
Disallow: /*&voice=
Disallow: /*?agent=
Disallow: /*&agent=
Disallow: /*?language=
Disallow: /*&language=
Disallow: /*?gender=
Disallow: /*&gender=
Disallow: /*?sort=
Disallow: /*&sort=
Disallow: /*?new=
Disallow: /*&new=
Disallow: /*?is_free=
Disallow: /*&is_free=
Disallow: /*?tool=
Disallow: /*&tool=
Disallow: /*?from=
Disallow: /*&from=
Disallow: /*?next=
Disallow: /*&next=