Synthetic teaching example for a website delivery brief. Not a client result, live account test or certification.
Keep a confidential preview private, not merely out of a report
Before a new website preview contains customer material, agree who can open it. A noindex directive concerns search indexing, not whether a person with the URL can read the page. A robots exclusion concerns crawling and can prevent a crawler from seeing the noindex instruction at all.
| Synthetic environment | Declared controls | Finite interpretation |
|---|---|---|
| A | Public, robots blocked, noindex | Not private; crawler blocked from inspecting the directive |
| B | Public, crawl allowed, noindex | Not private; directive supplied but live access still needs verification |
| C | Authentication required | Declared private control, not tested by this fixture |
Use the filled example
Download the finite Node checker and filled input. Run with Node 24 or newer; no account, API or network is used.
node checks.mjs staging-indexing-boundary staging-indexing-boundary.jsonThe checker reports privateAccessControl and noindexCanBeObserved separately. It deliberately does not call C “secure”: authentication:true is an input declaration, not proof of correct access control. Similarly, a supplied noindex value does not show a particular search engine fetched the page or removed a previously known URL.
An acceptance brief with two separate questions
First, verify confidentiality through authorised and unauthorised requests, asset access, share-link behaviour and caches. Do not place secrets or customer records in a public preview while waiting for search settings. Second, inspect indexing instructions on the actual response, their visibility to the intended crawler and any agreed preview-to-production transition.
Before launch, compare the new production hostname and intended indexing state with the approved checklist. A staging restriction accidentally copied to production is a different failure from a public confidential preview. Give each decision an owner and retain redacted response evidence without exposing access tokens.
This no-network worksheet changes no preview settings, old robots.txt, authentication, canonical or live page. It is not a security audit or a removal request. Use actual permitted response checks before relying on a declaration, and keep any old-site change outside an additive publication contract unless separately authorised.
Primary facts and limits
Google noindex guidance: A crawler must be able to access the page to see noindex; placing noindex in robots.txt is unsupported.
Sources reviewed 15 October 2026. The supplied input and acceptance rules are original teaching design; check current platform versions and your actual delivery contract before implementation.