<meta name="robots">
A request for discretion to the indexers — which the polite ones honor.
The per-page counterpart of robots.txt: noindex asks to stay out of the index, nofollow asks that the links there not be followed, and values like noarchive tune the rest. It is a request, not a lock — an ill-mannered bot ignores it.
The classic myth is stacking both defenses: blocking the page in robots.txt and declaring noindex. If the crawler cannot read the page, it never sees the request — and the page may stay in the index on the strength of other people's links. To vanish from the catalogue, the page must be readable.
<meta name="robots" content="noindex, nofollow">