Published April 7, 2026 · Updated April 7, 2026
CVE-2026-35020 is a high-severity OS command injection vulnerability found in Anthropic's Claude Code CLI and Claude Agent SDK. Local attackers can exploit this flaw by manipulating the TERMINAL environment variable to execute arbitrary commands, which can lead to system compromise. To mitigate this risk, ensure that the CLI and SDK are configured properly and monitor for unusual command execution patterns in your environment. Update to any available patched versions from Anthropic.
Anthropic Claude Code CLI and Claude Agent SDK contain an OS command injection vulnerability in the command lookup helper and deep-link terminal launcher that allows local attackers to execute arbitrary commands by manipulating the TERMINAL environment variable. Attackers can inject shell metacharacters into the TERMINAL variable which are interpreted by /bin/sh when the command lookup helper constructs and executes shell commands with shell=true. The vulnerability can be triggered during normal CLI execution as well as via the deep-link handler path, resulting in arbitrary command execution with the privileges of the user running the CLI.
A high-severity OS command injection vulnerability (CVE-2026-35020) in Anthropic's Claude Code CLI and Claude Agent SDK allows local exploit through environment variable manipulation. Immediate remediation is crucial.