Check the page plan against what is already live
Existing Site Guard asks a question a CSV cannot answer on its own: will this new page collide with a URL or page that already exists?
Last updated
What is checked
The scan retrieves a bounded view of the site and compares planned page records with pages it actually retrieved.
- Existing URL collision — Blocked.
- Existing title collision — Review.
- Existing body near-duplicate — Blocked.
- Existing intent collision — only when a connected source actually provides target intent.
How the crawl is bounded
The public scan is deliberately constrained. It follows robots.txt, keeps redirects on the requested origin, caps requests and limits concurrency.
- Default request budget: 45 including sitemap, pages and redirect hops.
- Real crawl concurrency: 3.
- The scanner rejects its own host and non-public/private network targets.
Coverage is part of the result
A scan can only compare pages it actually retrieved. Pages blocked by robots, outside the request budget, missing body text or lacking target-query data are not silently treated as clean.
- existingCount tells you how many live pages were retrieved.
- existingWithBodyCount tells you how many could be compared for body similarity.
- intentMeasured is false when the live source did not expose target intent.
What to do with a collision
A collision is a publishing decision, not merely a warning. Reuse, update, merge or differentiate the existing page before creating another URL that competes with it.
Next action
Use the rule reference when a page needs a decision, or continue through the Docs in order.