Host code execution via model-controlled sandbox override
Published Jun 2, 2026 · Updated Jun 2, 2026
Sandbox bypass in Gitlawb OpenClaude before 0.5.1 allows remote attackers to execute arbitrary commands on the host via prompt injection. The BashTool schema exposes dangerouslyDisableSandbox to model output, and shouldUseSandbox() skips confinement when that flag meets the default-true unsandboxed-command policy. Exploitation requires attacker-controlled content to influence model output while sandboxing is enabled; resulting commands run with the OpenClaude process's host privileges.
Summary
What happened
Sandbox bypass in Gitlawb OpenClaude before 0.5.1 allows remote attackers to execute arbitrary commands on the host via prompt injection. The BashTool schema exposes dangerouslyDisableSandbox to model output, and shouldUseSandbox() skips confinement when that flag meets the default-true unsandboxed-command policy. Exploitation requires attacker-controlled content to influence model output while sandboxing is enabled; resulting commands run with the OpenClaude process's host privileges.
The record
- CVE
- CVE-2026-42074
- Published
- Jun 2, 2026
- Updated
- Jun 2, 2026
- Vendor
- Gitlawb
- Product
- OpenClaude
- Classifications
- CWE-284, CWE-306, T1059
- Attack vector
- network
- Privileges
- unauthenticated
Timeline
How it unfolded
- Jun 2, 2026CVE publishedPublication date reported by the CVE source.
- Jun 2, 2026Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< 0.5.1
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
- GHSA-m77w-p5jj-xmhg proof of conceptproof of concept · validated
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo