CVE-2026-28471: Identity Impersonation in Matrix Plugin for OpenClaw

The core of this vulnerability lies in the DM allowlist matching process within the Matrix plugin. Instead of validating the sender's identity against the homeserver, the plugin allows matching against display names and localparts without proper verification. An attacker can exploit this by crafting Matrix messages with display names or localparts that exactly match entries in the allowlist, effectively bypassing the intended security controls. This allows the attacker to impersonate legitimate users, potentially gaining access to their private messages, initiating actions on their behalf, or disrupting communication flows within the OpenClaw environment. The potential blast radius depends on the sensitivity of the data handled within the DM and the permissions granted to the impersonated user.

1. Reserved2026-02-27
2. Published2026-03-05
3. Modified2026-03-09
4. EPSS updated2026-03-23

The primary mitigation for CVE-2026-28471 is to upgrade OpenClaw to version 2026.2.2 or later, which includes the fix for this vulnerability. If an immediate upgrade is not feasible due to compatibility issues or downtime constraints, consider implementing temporary workarounds. While a direct WAF rule is unlikely to be effective due to the nature of the vulnerability, carefully reviewing and tightening the DM allowlist rules to prevent overly permissive matching can reduce the attack surface. Monitor Matrix logs for suspicious activity, particularly messages originating from unexpected homeservers or with unusual display names. After upgrading, confirm the fix by attempting to send a Matrix message with a display name that should be blocked by the allowlist; the message should be rejected.