External risk intelligence

OpenClaw Dashboard Stored XSS via Crafted Username in Failed Login

CVE advisorySeverity: CRITICAL (CVSS 9.3)

CVE-2026-66418

The vulnerability exists in a dashboard product designed for administrative monitoring and management. Such dashboards are commonly deployed as internet-facing or externally reachable web applications to allow administrators to monitor systems remotely, making the login and management interfaces accessible from the network.

Cross-site Scripting

Halo Surface Signal: 4 out of 5 — likely to be public-facing.

External exposure likelihood

Horizon Alert

Summary of the vulnerability and why it matters

This vulnerability affects a dashboard application that administrators use to manage systems. An attacker can inject malicious code through a failed login attempt, which could then execute with administrative privileges when an administrator views logs. The primary concern is to confirm if this specific dashboard is deployed within your environment and assess any potential exposure.

  • Attackers can inject malicious code via failed logins.
  • Enables unauthorized access to administrative functions.
  • Confirm relevance and exposure within your environment.

Attack Path

How an attacker could exploit the issue

An attacker can reach this vulnerability by first sending a crafted username during a failed login attempt to a web-based dashboard. This username is stored in the system's audit log. When an administrator views the notification panel, the malicious username is displayed without proper escaping, allowing injected HTML and script to run within the administrator's browser session. This can lead to the attacker being able to edit agent instruction files or change system configurations.

  • No authentication needed to trigger.
  • Crafted username in failed login POST.
  • Execute code in administrator session.

Live Threat

Current exploitation, exposure, and threat context

This vulnerability could allow an unauthenticated remote attacker to execute arbitrary code within an administrator's session. This is possible when an attacker crafts a malicious username submitted during a failed login attempt. The username is then stored in the audit log and rendered in the notification panel, leading to the execution of injected HTML and script payloads in the administrator's browser context. This, in turn, could allow the attacker to interact with authenticated administrative functions.

  • Administrator session data and authorized actions.
  • Crafted username in failed login POST request.
  • Unauthorized administrative control or changes.

Operational Fix

Recommended remediation, mitigation, and detection steps

Teams responsible for administrative dashboards and their underlying infrastructure should prioritize this vulnerability. The first step is to locate all instances of OpenClaw Dashboard, determine their exposure (internal vs. external), and identify the accountable system owner or application owner. Once confirmed as a business-critical or externally accessible asset, a risk-based remediation plan can be developed, potentially involving vendor coordination or temporary risk reduction measures.

  • Identify asset owners and scope.
  • Verify administrator access and reachability.
  • Plan remediation based on risk.

Supplementary metadata

Validate whether this threat affects your internet-facing exposure.

Halo Threat Intelligence helps prioritize remediation with Halo Surface Signal and H/A/L/O context. Start exposure validation with a free external attack surface trial.

Frequently asked questions

What is OpenClaw Dashboard?

OpenClaw Dashboard is a web-based administrative management tool. It provides a centralized interface for monitoring system health and managing infrastructure configurations. Administrators rely on it to oversee system status and perform routine maintenance tasks through a graphical notification panel.

What does CWE-79 mean for CVE-2026-66418?

CWE-79 refers to Cross-Site Scripting (XSS). In this CVE, the dashboard fails to sanitize user input, allowing an attacker to inject harmful scripts into the system. When an administrator views the audit logs, their browser treats this injected text as trusted code, executing it within the administrator's active session and potentially granting unauthorized control over the dashboard.

How is this vulnerability triggered?

An attacker triggers this by submitting a specially crafted username during a failed login attempt. The dashboard stores this malicious string in its audit log. Importantly, successful logins do not trigger this, nor does simply viewing the login page; the malicious script only executes when an administrator subsequently opens the notification panel where those logs are rendered.

Is my instance at risk?

Halo Surface Signal indicates this is a likely risk for many organizations. Because OpenClaw Dashboard is designed for remote monitoring, these interfaces are frequently deployed as internet-facing web applications. If your dashboard is accessible from outside your internal network, it is exposed to remote, unauthenticated attackers.

What should I do first to respond?

Begin by auditing your network to locate all active OpenClaw Dashboard installations. Confirm which instances are reachable from the internet versus those restricted to internal access. Identify the system owners for each instance and prioritize remediation for externally accessible assets, as they are the most vulnerable to unauthenticated exploitation.

References