What’s Vulnerable › Vulnerabilities

CVE-2026-42074: OpenClaude: Sandbox Bypass via Model-Controlled `dangerouslyDisableSandbox` Input

• Severity: critical (CVSS 9.8 v3.1)
• Exploited in the wild: not listed in CISA KEV
• EPSS: 0.6% probability of exploitation in the next 30 days
• Public exploit or detection template: yes
• Published: 2026-06-02 · updated 2026-07-22
• Weakness: CWE-284

Description

OpenClaude is an open-source coding-agent command line interface for cloud and local model providers. Prior to version 0.5.1, the dangerouslyDisableSandbox parameter is exposed as part of the BashTool input schema, meaning the LLM (an untrusted principal per the project's own threat model) can set it to true in any tool_use response. Combined with the default allowUnsandboxedCommands: true setting, a prompt-injected model can escape the sandbox for any arbitrary command, achieving full host-level code execution. This issue has been patched in version 0.5.1.

Affected products

• Gitlawb openclaude: before 0.5.1 — fixed in 0.5.1

How to fix

Upgrade to a fixed version: Gitlawb openclaude 0.5.1. Check the vendor advisory for mitigations.

References

• https://github.com/Gitlawb/openclaude/commit/aab489055c53dd64369414116fe93226d2656273
• https://github.com/Gitlawb/openclaude/pull/778
• https://github.com/Gitlawb/openclaude/security/advisories/GHSA-m77w-p5jj-xmhg
• https://github.com/Gitlawb/openclaude/security/advisories/GHSA-m77w-p5jj-xmhg

Get alerted when CVE-2026-42074 or the next vulnerability affects your stack