Use this guide for a published auxiliary page, such as a search-results page, that should remain directly accessible without appearing in indexes. These settings are independent; choose the ones that match your publishing setup.
Excluding a page is not access control. Its public URL still works. These settings do not make private information safe to publish.
Edit the existing page
Inside the page’s current front matter, add or update these fields while keeping its title, layout and permalink:
robots: "noindex, follow"
sitemap: false
include_on_search: false
robots asks external crawlers not to index the page. Crawlers need to access
it to read the instruction; do not block it in robots.txt and assume they can
still read noindex. Removal from existing search results is not immediate.
sitemap: false excludes it from jekyll-sitemap if that plugin generates your
sitemap. It does not itself prevent external indexing. include_on_search: false
excludes it from the built-in Lunr, Fuse.js and Simple-Jekyll-Search indexes.
For Algolia, configure its files_to_exclude for the source file and upload the updated index using the Algolia guide. The local-search flag is not an Algolia exclusion. Google Custom Search uses external indexing rather than these local index files. Other sitemap/index tools need their own configuration.
Publish and verify
Commit on the publishing branch, wait for Actions to succeed and open the
page directly. Check its content still loads and its HTML source contains the
noindex, follow robots value. With jekyll-sitemap, check the page’s actual URL
is absent from the published sitemap.xml.
For local search, search for a distinctive phrase from the page after rebuilding and check that the page no longer appears. For Algolia, verify the updated uploaded records separately. The documentation’s Search page uses these three fields; its source is a real configuration example, not evidence of immediate removal by a crawler.
If the result is unexpected, check explicit front matter and more-specific configuration defaults, the actual search provider and the latest publishing run. Use the field constraints and publishing checks. Do not assume tags, archives, error pages or pagination are automatically excluded by their layout or path.