Skip to content

Claude Code

Claude Code v2.1.149 released

Written by an AI model from the linked sources and checked against them by a second model. How we work

All about: Claude Code versions and changes →

The update revamps the /usage command to show a detailed breakdown of limits by category: spend is now allocated across skills, subagents, plugins, and individual MCP servers. This simplifies budget control on complex tasks.

The release also closes a PowerShell vulnerability allowing stealth directory traversal. A critical Bash find utility bug that caused host crashes on macOS with large directory trees is fixed.

Additionally, the team improved keyboard navigation in /diff, added Markdown checkbox support, and resolved terminal freezing on config approval.

Anthropic

Anthropic co-founder on the risk of people turning into zombies

Anthropic co-founder Jack Clarke warns that over-reliance on AI advice like Claude’s can turn people into ‘philosophical zombies’ who act correctly but lose human agency and the ability to learn from mistakes. He suggests AI systems should encourage users to take breaks and discuss issues with real people.

No comments yet. Be the first.

Join the conversation

Sign in to leave a comment. It appears after moderation.