Openclaw cli: OpenClaw: QQBot pre-dispatch slash commands could skip allowFrom checks (CVE-2026-53834)

CVE-2026-53834 describes a security flaw in OpenClaw CLI where QQBot pre-dispatch slash commands can skip the allowFrom checks designed to restrict command invocation sources. This allows a sender capable of invoking slash commands to dispatch commands before the allowFrom policy is enforced, potentially enabling command handling from unauthorized senders if the feature is enabled and reachable. The vulnerability is scoped to the QQBot slash command feature and does not affect the overall trusted-operator security model. The issue is fixed in version 2026.4.27.

If the vulnerable feature is enabled and accessible, attackers or lower-trust senders could execute slash commands that should have been blocked by allowFrom policies. The practical impact depends on the operator's configuration, including how narrowly allowlists are defined and whether untrusted users share the same Gateway. The vulnerability could lead to unauthorized command execution, impacting integrity but not confidentiality or availability.

A patch is available in OpenClaw CLI version 2026.4.27. Operators should upgrade to this version to remediate the vulnerability. Until patched, restrict exposure of QQBot slash commands. Additionally, maintain narrow channel and tool allowlists, avoid sharing Gateways between mutually untrusted users, and disable the affected QQBot slash command feature if it is not needed.