No-op
emergingReal practice, thin evidence. Mostly vendor-stated or reported rather than measured.
An instruction the model already obeys by default. It pays load to say nothing. The test — does this change behaviour versus the default? — is model-relative rather than reader-relative, so two people arguing about a no-op are arguing about the default and settle it by running the document, not by debating it. When a sentence fails the test, delete the sentence rather than trimming words out of it. Related and often confused: predictable means the agent takes the same process every run, not that it produces the same output.