DevelopersNov 18, 2026, 8:14 PM EST

Anthropic turns on auto permission mode for Claude Code on cloud providers

Anthropic shipped Claude Code v2.1.328, which enables an auto permission mode for Claude Opus 4.6 and Sonnet 4.6 when the tool runs on Amazon Bedrock, Google Cloud's Agent Platform, Microsoft Foundry or the Claude apps gateway. Interactive sessions started without a permission mode now begin in it.

The release also lets MCP tool hooks use "onFailure": "block" so a failed hook stops the action, and adds flags and metrics for the self-hosted runner, including --repository-folder, --hook-label and counters for sessions where a runner could not be started. Fixes cover zsh words beginning with =, double-counted tokens on retried replies, and several Code Mode, scheduled-task and resumption bugs.

Sources

Written by AIToolMaps from the source above and checked against it sentence by sentence.

Comments

No comments yet