User-agent: *
Allow: /

# Block all parameter-based URLs (filters, sort, pagination)
Disallow: /*?

# AI assistants. Same policy as everyone else — these groups exist so the policy can be tuned per
# agent without touching the wildcard group, and so it is obvious which agents we considered.
#
# ⚠️ robots.txt groups do NOT inherit. An agent named below obeys ONLY its own group, so every
# group must repeat `Disallow: /*?` — dropping that line silently opens parameterised URLs to
# exactly the agents we most care about. See docs/AI_TRAFFIC.md.

# User-triggered fetchers — dispatched when a human asks an assistant about a page. Blocking these
# makes the assistant tell the user it could not read our site.
User-agent: ChatGPT-User
Allow: /
Disallow: /*?

User-agent: Claude-User
Allow: /
Disallow: /*?

User-agent: Perplexity-User
Allow: /
Disallow: /*?

User-agent: Google-Agent
Allow: /
Disallow: /*?

# Search indexers — build the citation index AI answers link from. Blocking these removes us from
# AI answer citations while leaving model training unaffected.
User-agent: OAI-SearchBot
Allow: /
Disallow: /*?

User-agent: Claude-SearchBot
Allow: /
Disallow: /*?

User-agent: PerplexityBot
Allow: /
Disallow: /*?

# Training crawlers (GPTBot, ClaudeBot, Bytespider, Meta-ExternalAgent) are intentionally NOT
# listed: they fall through to the wildcard group above and stay allowed. Blocking them costs no
# live AI visibility, so it is a content-licensing decision, not a technical one — take it
# deliberately rather than by omission.

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