Claude Code hits $2.5B in revenue and ships auto mode, an AI classifier that decides what's safe to run on your machine

About this article

Anthropic dropped three features for Claude Code on Monday, but the interesting one is auto mode. Until now you had two choices: approve every single file write and bash command manually, or use the skip permissions flag which lets the AI do whatever it wants. Auto mode sits in the middle. An AI safety classifier evaluates every tool call in real time. Routine stuff like writing files and running tests gets auto approved. Destructive operations like mass deletion or data exfiltration get bloc...