In auto mode a classifier checks actions such as shell commands and network requests before they run. Anthropic's documentation says that from 2.1.278 Claude Code asks the server to perform those checks as part of the session's own model…
In the News: September 19, 2026 (Extra 2)
Claude Code 2.1.278 stops charging for auto mode's safety classifier, unless a gateway between you and the API is rewriting the traffic.
Extra edition
Machine-readable
Download Markdown