Loading vulnerability…
Privilege Escalation in OpenClaw allowFrom Discord Identity Validation
CVE-2026-53849 is a privilege escalation vulnerability affecting OpenClaw versions before 2026.5.7. The flaw is in the allowFrom access-control feature, which authorizes Discord-linked identities using mutable display names rather than immutable Discord user IDs. Because display names can be changed by the account holder, an attacker with a Discord account can rename their account to match an authorized policy entry and be incorrectly treated as that trusted identity. This results in improper identity validation and authorization bypass within Discord-integrated access control, allowing unauthorized agent access intended for another user.
What it means and what to do now: patch path, interim mitigations, and the assume-compromise checklist.
What an attacker gets, and what they’ve been doing with it.
Successful exploitation allows an unauthorized Discord user to impersonate an allowed identity for the purpose of OpenClaw authorization. This can lead to privilege escalation and unauthorized access to agent functionality that should be restricted to a different Discord account. The practical impact depends on how the deployment uses the affected feature, but it can expose protected agent capabilities to lower-trust users remotely.
If you can’t patch tonight, do this now.
If immediate patching is not possible, avoid using Discord display names for access-control decisions and restrict authorization to immutable Discord user IDs only. Disable the affected allowFrom behavior where feasible. As additional hardening, keep channel and tool allowlists narrow and avoid sharing a single gateway or trust boundary among mutually untrusted users.
Patch, then assume compromise.
Upgrade OpenClaw to version 2026.5.7 or later. Authorization logic should be corrected to bind policy decisions to immutable Discord user IDs rather than mutable display names. Any existing allowlists or policy entries based on display-name matching should be reviewed and migrated to stable account identifiers.
No public exploits tracked yet. Mallory keeps watching.
Products and vendors Mallory has correlated with this vulnerability. Open in Mallory to drill down to specific CPE configurations and version ranges.
5 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.