# robots.txt — zaps.design

User-agent: *
# Content Signals (contentsignals.org): search and grounding are welcome, model
# training is not. Stated as an express reservation of rights, not a hint.
Content-Signal: ai-train=no, search=yes, ai-input=yes
Disallow: /login
Disallow: /auth/
Disallow: /*/edit

# Query params with no SEO value — don't burn crawl budget
Disallow: /*?*utm_
Disallow: /*?*fbclid=
Disallow: /*?*gclid=

# For agents. robots.txt is the one file every crawler fetches, so it is the most
# reliable place to say where the machine-readable surface lives — the well-known
# documents are not listable and are deliberately unindexed, which leaves an agent
# with no way to find them but to be told.
#   Skill:       https://zaps.design/.well-known/agent-skills/zaps-design/SKILL.md
#   API docs:    https://zaps.design/docs/api
#   MCP server:  https://api.zaps.design/mcp
# Llms-Txt is not a registered directive; it is a comment-adjacent hint for the
# crawlers that read one, and the Link headers carry the same targets for those
# that do not.
Llms-Txt: https://zaps.design/llms.txt

# /sitemap.xml is a sitemap index → sitemap-marketing.xml (static),
# sitemap-app.xml (app routes), sitemap-collections.xml (API-driven).
Sitemap: https://zaps.design/sitemap.xml