Vulnerability · Published Jun 2, 2026 · Assigned by GitHub_M

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

Critical 9.3

In plain English

A critical-severity missing authentication in openclaude from Gitlawb. Proof-of-concept exploit code is public. The chance it is exploited in the next 30 days is estimated at 0.6% (higher than 48% of all known vulnerabilities). Fixed versions are available - upgrade the affected package.

Severity (CVSS 4.0)9.3Critical

Exploit odds (30 days)0.6%top 51.7% of all CVEs

Exploited in the wildNo reportsCISA SSVC: poc

Attack toolingNoneno public tooling known

Official 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 packages & fixed versions

Ecosystem	Package	Fixed in	Advisory
npm	openclaude	0.5.1	GHSA-m77w-p5jj-xmhg

Affected products (as reported)

Vendor	Product	Versions
Gitlawb	openclaude	< 0.5.1

Weakness: CWE-306, CWE-284 · Vector: CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N

References

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