Disclosed weakness · no CVE recorded

OpenClaw could attach private device images to a chat reply

OpenClaw · OpenClaw · paired-node media

First documented disclosure · Review checkpoint

Summary

An agent inspecting a paired device could send its image to a chat, including a group, when the requested answer was text only.

Severity and exploitation evidence

No CVSS score recorded in this review.

Exploitation not established: The upstream sources document the behavior and correction. They identify no malicious use.

Exploitation evidence source

Technical severity is not a measure of observed attacks or of risk to your deployment. Attempts, demonstrations, and successful compromise retain the qualifications in the source.

How the weakness works

The reply system treated inspected images as attachments authorized for delivery.

Conditions required

A session with paired-device access inspects a photo, camera image or screen.

Potential impact

The recipient chat could receive visual information the operator intended only for inspection.

Affected versions and configurations

The upstream report identifies the affected behavior without specifying a complete version range.

Fix and mitigation status

The 2026.11.4 stable release includes this issue-specific correction.

Practical next steps

Deploy the correction and review prior chat attachments where devices handled sensitive information.

Operating lesson

Require a separate decision to send media an assistant can inspect.

Evidence and limits

Earlier browser screenshot reports concern another tool. The paired-device disclosure is dated separately.

Sources