Follow Discord
Sweep 30 Oct 2026 · 01:02Z Build v2.1.314 532 read Stable v2.1.308 Latest v2.1.314 Next v2.1.314 Feeds RSS JSON llms.txt llms-full.txt Unofficial

Claude Code v2.1.314 ·

Auto mode block notices are reworded, grouped and screened for sensitive text

Auto mode now says an action was "blocked", folds repeated blocks into one notice, and blanks out reasons that look sensitive

You'll notice Improvements
JSON All of v2.1.314
You'll noticeTier: how much it should matter to you
2Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
Auto ModeArea: what it touches
ImprovementsKind: in v2.1.314,
ImprovementsSection of the release
What

Auto mode is a permission mode where Claude Code decides on its own whether an action can go ahead. When it stops an action, it shows a warning notice. That notice has changed in several ways:

  • The wording is now "blocked by auto mode" instead of "denied by auto mode".
  • Several blocks in a short space of time are folded into one notice that reads "N recent blocks by auto mode", instead of one notice per block.
  • The reason for the block is checked for sensitive content. If the check matches, the reason, the displayed text and the blocked action's input are blanked out.
  • The reason is shortened for display.
  • The " · /permissions" hint at the end is left off in some cases.

The prompt auto mode shows when it wants you to decide also changed. It now reads "Auto mode needs your confirmation for this" instead of "Auto mode classifier requires confirmation for this".

Why

In auto mode you will see fewer, plainer block notices, and text that may be sensitive is screened before it is shown or stored.

How sure we are
One source agreesOne thing we can check says the same as this entry.
Anthropic's release notes agreeFixed auto mode reporting a call as blocked, and listing it in /permissions, when its safety check gave no verdict

See this entry in the whole of v2.1.314 →

Feedback