Executive brief
OpenClaw, a tool used for managing communication gateways and agent bindings, contains a security flaw in its Feishu integration. Authenticated users can bypass intended security restrictions to create or modify agent bindings that should normally be protected. This could allow an unauthorized user to change how messages are routed or handled, potentially leading to unauthorized configuration changes within the communication platform.
Technical details
A configuration enforcement bypass vulnerability exists in OpenClaw's Feishu dynamic-agent bindings due to incorrect authorization (CWE-863). The root cause is a failure to honor 'config-write' controls when a Feishu sender utilizes the dynamic-agent binding feature. An authenticated attacker with low privileges can exploit this over the network to create or update bindings, effectively modifying the sender-agent binding state beyond the intended security policy. This allows for unauthorized configuration modifications within the affected component. The issue is resolved in version 2026.5.6; users can mitigate the risk by disabling sender-created Feishu dynamic-agent bindings.
Affected products
- OpenClaw openclaw < 2026.5.6
Timeline
- 2026-05-28: advisory: GitHub Security Advisory GHSA-3wqp-prf6-2m72 published
- 2026-06-12: disclosed: CVE-2026-53835 published
- 2026-05-28: patched: Version 2026.5.6 released