Source Intelligence

DisclaimerUnofficial, and not affiliated with Anthropic. Nearly all of this is read straight out of what ships: npm bundles, captured prompts, published docs. Anthropic's own notes go in verbatim, marked as theirs. The rest is my reading, and every entry carries the strings behind it. If one looks wrong, vote it down and say why.

All of v2.1.72 Home All releases olderv2.1.71 v2.1.73newer
Claude Code v2.1.72

Auto Mode Opt-In Dialog

What

When launching Claude with --mode auto, users who haven't previously opted in now see a consent dialog explaining what auto mode does.

Details
  • Dialog text: "Auto mode lets Claude run commands without prompting — each action is first checked for safety. Claude can still make mistakes; use in isolated environments."
  • Declining the dialog exits the application
  • Once accepted, the opt-in is remembered
  • New "Yes, and use auto mode" and "Yes, clear context" options in the context-clearing prompt
Evidence

New dialog component (search for "AutoModeOptInDialog")

Strings lifted out of the shipped bundle, so the claim above can be checked against them.

See this entry in the whole of v2.1.72 →