Has everything been described?

Understudy::nothingElse($repository);                    // every call was accounted for
Understudy::nothingElse($repository, $clock, $mailer);   // across several doubles
Understudy::allVerified($repository);                    // expectations met AND nothing else

What counts as accounted for

	Accounts for a call
An expect() that matched it	yes
A successful verify() that claimed it	yes
A failed verify()	no
A when() stub	no

A stub accounts for nothing because it is permission, not a description of what happened. This is the rule that makes nothingElse() worth having: a test can stub freely and still assert that the subject made no call it did not mean to.

Why one line, many doubles

nothingElse() takes any number of doubles, so one line closes out the whole test. A failure names every offender rather than stopping at the first — the second unexpected call is usually as interesting as the first, and finding it one run at a time is the slow way.

The trap this closes

A spy-style test that counts calls will happily pass while the subject also makes a call with entirely different arguments:

expect(fn () => $repository->save($expected));
// subject also calls $repository->save($somethingElse) — still green

expect() claims the calls it matched. It says nothing about the rest. Add nothingElse(), or use allVerified(), which is both checks in one:

Understudy::allVerified($repository);

allVerified() checks ordered expectations too.

See the cookbook case for the full worked example — it is the trap people carry over from Mockery most often.