Skip to content

CVE-2025-66032 Claude Code Command Validation Bypass Allows Arbitrary Code Execution

HighCVSS 8.7 EPSS 0.74%
Published
Updated
Exploit status
No public exploit known

Description

Claude Code is an agentic coding tool. Prior to 1.0.93, Due to errors in parsing shell commands related to $IFS and short CLI flags, it was possible to bypass the Claude Code read-only validation and trigger arbitrary code execution. Reliably exploiting this requires the ability to add untrusted content into a Claude Code context window. This vulnerability is fixed in 1.0.93.

Affected products

ProductAffected versionsFixed in
anthropic-ai @anthropic-ai/claude-code(npm) < 1.0.93 1.0.93
anthropics claude-code = < 1.0.93 No fix listed
Anthropic claude code < 1.0.93 1.0.93

References