# How automated systems may use our content: search results, live AI answers
# and AI training are all welcome. (Was a contentsignals.org "Content-Signal:
# search=yes, ai-input=yes, ai-train=yes" line until Oct 2026; Lighthouse flags
# it as an unknown directive, and all-yes is the default anyway.)
User-agent: *
Allow: /
Disallow: /admin
Disallow: /admin/
Disallow: /account
Disallow: /account/
Disallow: /api/

# Tool job status + result URLs are user-specific (6-month TTL). The tool
# landing pages stay indexable; the per-job pages don't add crawl value.
Disallow: /tools/*/jobs/

# Mark detail pages are for people using the tools (one per registration —
# millions of them). Never crawl or index them; the official record is the
# canonical source.
Disallow: /mark/

# Status JSON is for uptime monitors, not search
Disallow: /status.json

# The assistant plugin endpoint speaks JSON to machines, not to readers
Disallow: /mcp

# Demo recordings sent to plugin reviewers
Disallow: /demo

# Canada search results (?q=, with status and class filters) are noindex and
# multiply into thousands of filter combinations; the search page itself stays.
Disallow: /canada-trademark-search?

# Auth pages — no point indexing login/register flows
Disallow: /login
Disallow: /no-track
Disallow: /register
Disallow: /forgot
Disallow: /reset-password

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