Claude Code Command Validation Bypass Allows Arbitrary Code Execution

Claude Code before version 1.0.93 has a flaw that allows attackers to run arbitrary code by bypassing safety checks through specially crafted shell commands. This matters because it could let someone execute dangerous commands on a system if they can inject malicious content into Claude Code.

The vulnerability exists in command parsing logic that fails to properly validate shell metacharacters (specifically $IFS and short CLI flags), allowing attackers to circumvent read-only restrictions through command injection. Exploitation requires ability to inject untrusted content into the Claude Code context; successful exploitation results in arbitrary code execution with the privileges of the running process.