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

GHSA-6c4r-g249-wv3c: OpenClaw: Sandboxed session spawn could expose the real workspace path to child prompts

mediumCVSS 4.0 5.3

Summary Sandboxed session spawn could expose the real workspace path to child prompts. In affected versions, a child session spawned from a sandboxed parent could forward the host workspace path into the child session prompt. This advisory is scoped to the named feature and configuration.

Published Jul 2, 2026

Affected packages

openclaw

Ecosystem
npm
Affected
<= 2026.4.25
Fixed in
2026.4.26
Check
npm ls openclaw

Advisory details

From the GitHub Advisory Database (CC BY 4.0).

Summary

Sandboxed session spawn could expose the real workspace path to child prompts. In affected versions, a child session spawned from a sandboxed parent could forward the host workspace path into the child session prompt.

This advisory is scoped to the named feature and configuration. It does not change OpenClaw's trusted-operator model: authenticated Gateway operators, installed plugins, and intentional local execution surfaces remain trusted unless a separate policy, approval, allowlist, sandbox, or auth boundary is crossed.

Impact

When the affected feature is enabled and reachable, this could reveal host workspace location or related memory context to the child model. Practical impact depends on the operator's configuration and whether lower-trust input can reach that path.

Patched Versions

The first stable patched version is 2026.4.26.

Mitigations

avoid spawning child sessions from sensitive sandboxed workspaces until patched. As general hardening, keep channel and tool allowlists narrow, avoid sharing one Gateway between mutually untrusted users, and disable the affected feature when it is not needed.

Credited to anshumanbh.

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 OpenClaw vulnerabilities

All OpenClaw entries