GHSA-6c4r-g249-wv3cMedium▾ SunlitOpenClaw: Sandboxed session spawn could expose the real workspace path to child prompts
▾ Sunlit zone — Low / medium · no exploitation signal
impact 27.5 · likelihood 0 · exploitation 0
Need a working PoC? Pro members can cast a request and our team develops one — it lands right here.
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.
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.
The first stable patched version is 2026.4.26.
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.
openclaw <= 2026.4.25Upgrade to a patched release:
openclaw 2026.4.26Connected by shared product, vendor, weakness, or advisory.
CVE-2026-35630High· 8.0OpenClaw: QQBot native approval buttons did not enforce configured approver identity
GHSA-c29c-2q9c-pc86HighOpenClaw: Slack allowFrom could bind to mutable display names
GHSA-qjpc-qf9m-xwmrHigh· 8.8OpenClaw: Trusted-proxy Control UI WebSocket accepted client-declared scopes before pairing
GHSA-gp79-m99v-gjmhMediumOpenClaw: Mattermost handlers could fall open when channel type was missing
GHSA-w4v6-g3wm-w36cCriticalOpenClaw: QQBot admin commands could skip DM-only and allowFrom policy
GHSA-grc3-2j34-p6gmMediumOpenClaw: message.action forwarding could send Gateway credentials to model-supplied loopback URLs