Free AI Safety Workshop — in collaboration with explainx.aiSave your seat
AgentbeamBeta
AI agent vulnerability · Claude Code

CVE-2026-55607: Claude Code: Sandbox Escape via Git Worktree Path Confusion Allows Unsandboxed Code Execution

highCVSS 4.0 7.7GHSA-7835-87q9-rgvv

Claude Code's worktree handling allowed creation of worktrees named ".git" and navigation to worktrees outside the sandbox context, enabling git directory confusion attacks. By exploiting symlink manipulation and git fsmonitor execution during worktree operations, an attacker could overwrite files in the user's home directory (such as .zshenv), leading to code execution outside of seatbelt sandbox restrictions.

Published Jul 24, 2026

Affected packages

@anthropic-ai/claude-code

Ecosystem
npm
Affected
>= 2.1.38, < 2.1.163
Fixed in
2.1.163
Check
npm ls @anthropic-ai/claude-code

Advisory details

From the GitHub Advisory Database (CC BY 4.0).

Claude Code's worktree handling allowed creation of worktrees named ".git" and navigation to worktrees outside the sandbox context, enabling git directory confusion attacks. By exploiting symlink manipulation and git fsmonitor execution during worktree operations, an attacker could overwrite files in the user's home directory (such as .zshenv), leading to code execution outside of seatbelt sandbox restrictions. Reliably exploiting this required the user to clone a malicious repository containing prompt injection content and run Claude Code against it.

Users on standard Claude Code auto-update have received this fix automatically. Users performing manual updates are advised to update to the latest version.

Thank you to hackerone.com/metnew for reporting this issue.

References

How this page is made: the data panel comes from the GitHub Advisory Database, CISA’s KEV catalog, the NVD and FIRST EPSS, and updates twice a day. No analysis has been written for this entry yet. Methodology

More Claude Code vulnerabilities

All Claude Code entries