AnthropicGHSA-qxfv-fcpc-w36x
Command Injection in Claude Code rg command allowed bypass of user approval prompt for command execution
High8.7CVE-2025-58764 · Published Sep 9, 2025
Due to an error in command parsing, it was possible to bypass the Claude Code confirmation prompt to trigger execution of an untrusted command. Reliably exploiting this requires the ability to add untrusted content into a Claude Code context window. Users on standard Claude Code auto-update will have received this fix automatically. Users performing manual updates are advised to update to the latest version. Thank you to the NVIDIA AI Red Team for reporting this issue!
Affected versions
| Package | Affected | Fixed in |
|---|---|---|
| @anthropic-ai/claude-code npm | < 1.0.105 | 1.0.105 |
Details and references
- CVSS 4.0
- CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
- Severity from
- GitHub (reviewed advisory)
- Weakness
- CWE-94
More Anthropic advisories
All Anthropic| Date | Advisory | Severity | Fixed in |
|---|---|---|---|
| Oct 32025 | Command execution prior to Claude Code startup trust dialog | High8.7 | v1.0.111 |
| Oct 32025 | Permission deny bypass through symlink | Low2.3 | v1.0.120 |
| Sep 242025 | Claude Code Vulnerable to Arbitrary Code Execution via Plugin Autoloading with Specific Yarn Versions | High7.7 | v1.0.39 |
| Sep 92025 | Claude Code vulnerable to arbitrary code execution caused by maliciously configured git email | High8.7 | 1.0.105 |
| Sep 22025 | Claude Code Vulnerable to Arbitrary Code Execution Due to Insufficient Startup Warning | High8.7 | v1.0.87 |
| Aug 152025 | Permissive Default Allowlist Enables Unauthorized File Read and Network Exfiltration in Claude Code | High7.1 | v1.0.4 |