Anthropic now lets Claude users jump out of plane without parachute

Anthropic, prone to pronouncements about model safety, has given developers using Claude Code the opportunity to throw its guardrails out the window. The company's latest Claude Code build ( v2.1.311 ) supports an environmental variable designed to allow the AI agent to run the rm command, which deletes files, without asking for approval. The designated environmental variable (a value developers set in their shell) is called: CLAUDE_CODE_EVEN_MORE_DANGEROUSLY_SKIP_PERMISSIONS .