Claude Code v2.1.166: resilience and security
Anthropic released version 2.1.166 of the Claude Code CLI tool. The highlight is the fallbackModel setting, which lets users specify up to three backup models to activate in sequence if the primary model is overloaded or unavailable.
On the security front, inter-session messaging got more restrictive. Permission requests sent via SendMessage from other sessions no longer carry the user’s privileges, and auto-mode blocks them by default. The update also adds support for glob patterns in deny rules for tools.
Technical issues were fixed, including flickering terminals in JetBrains IDE version 2026.1+, PowerShell freezing on Windows, and a resource leak where claude --bg-pty-host processes could consume 100% CPU on macOS.
Up next
Anthropic doubles Claude Cowork usage limits, Video
Anthropic doubles Claude Cowork usage limits for all paid users through July 5. This lets users delegate more complex, multi-step tasks to the model.
Related stories
- Claude Code v2.1.163: version control and Bash fixes
- Claude Code update 2.1.161: bug fixes and new metrics
- Claude Code v2.1.160: security focus
- Claude Code 2.1.317: email check, SDK hook block, cost API
- Claude Code 2.1.314 adds write fence, IdP groups, compaction
- Claude Code 2.1.311 can drop its last rm -rf guard
Comments
No comments yet. Be the first.
Join the conversation
Sign in to leave a comment. It appears after moderation.