Anthropic has released Claude Code 2.1.325, which includes a security fix for zstd-compressed response decoding and significant changes to default permission behaviors for sandboxed Bash commands. Previously, Bash commands within the sandbox would execute automatically, but this update now requires explicit permission by default, potentially impacting unattended automation. Additionally, scheduled tasks and loops will pause before running skills that declare tool access or hooks, enhancing user control over potentially sensitive operations. AI
IMPACT Changes to default permissions in Claude Code may require adjustments for users relying on unattended automation with sandboxed Bash commands.
RANK_REASON This is a software update for a specific tool, not a frontier model release or significant industry event.
Read on dev.to — Claude Code tag →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →