Lobe Chat affected by Cross-Site Scripting(XSS) that can escalate to Remote Code Execution(RCE)

The core of the vulnerability lies in the Renderer component, which handles Mermaid diagram rendering. User-supplied content, originating from either user messages or AI-generated responses, is directly passed to the <Mermaid> component without proper sanitization. An attacker can craft a malicious Mermaid diagram containing JavaScript payloads. When this diagram is rendered, the injected script executes within the application's security context, allowing the attacker to potentially steal sensitive data, modify application behavior, or even gain complete control of the server. The potential for RCE significantly expands the attack surface and increases the severity of this vulnerability. This is particularly concerning in environments where @lobehub/chat is integrated with sensitive data or critical systems.

Applications utilizing @lobehub/chat versions prior to 2.0.0-next.180 are at significant risk. This includes chatbot applications, AI-powered assistants, and any system integrating @lobehub/chat for rendering Mermaid diagrams. Shared hosting environments where multiple applications share the same server instance are particularly vulnerable, as a compromise in one application could potentially impact others.

• nodejs / supply-chain:

Get-Process -Name node | Where-Object {$_.Path -match '@lobehub/chat'}

• nodejs / supply-chain:

Get-WinEvent -LogName Application -Filter "EventID=4688 AND SubjectUserName='@lobehub/chat'"

• generic web: Inspect HTTP requests and responses for unusual JavaScript code within Mermaid diagrams. Look for obfuscated or encoded scripts.

1. 2026-01-20Disclosure

   disclosure

1. 已保留2026-01-15
2. 发布日期2026-01-20
3. 修改日期2026-02-05
4. EPSS 更新日期2026-03-23

The primary mitigation for CVE-2026-23733 is to immediately upgrade to version 2.0.0-next.180 or later of @lobehub/chat. If upgrading is not immediately feasible, consider implementing temporary workarounds. Input validation and sanitization of Mermaid diagrams are crucial. Implement strict content security policies (CSP) to restrict the execution of inline scripts. Additionally, review and audit all user-generated content to identify and remove potentially malicious Mermaid diagrams. Monitor application logs for suspicious activity, particularly JavaScript execution attempts originating from unexpected sources. After upgrading, confirm the fix by attempting to render a known malicious Mermaid diagram and verifying that the script does not execute.