CVE-2026-23733: LobeChat Stored XSS Vulnerability

CVE-2026-23733 Overview

LobeChat, an open source chat application platform, contains a stored Cross-Site Scripting (XSS) vulnerability in its Mermaid artifact renderer that can be escalated to Remote Code Execution (RCE). The vulnerability exists in versions prior to 2.0.0-next.180 and allows attackers to execute arbitrary JavaScript within the application context. When exploited in the Electron desktop application, attackers can leverage the exposed electronAPI IPC bridge to run arbitrary system commands on the victim's machine.

Critical Impact
Stored XSS vulnerability can be escalated to full Remote Code Execution, allowing attackers to execute arbitrary system commands on victim machines through the Electron IPC bridge.

Affected Products

LobeChat versions prior to 2.0.0-next.180
LobeChat Electron desktop application with exposed electronAPI IPC bridge

Discovery Timeline

2026-01-18 - CVE-2026-23733 published to NVD
2026-01-18 - Last updated in NVD database

Technical Details for CVE-2026-23733

Vulnerability Analysis

This vulnerability is classified under CWE-94 (Improper Control of Generation of Code), representing a code injection flaw. The stored XSS vulnerability resides in the Mermaid artifact renderer component of LobeChat. Mermaid is a JavaScript-based diagramming and charting tool that renders markdown-inspired text definitions to create diagrams dynamically. The renderer fails to properly sanitize user-controlled input before processing, allowing malicious JavaScript to be stored and later executed when other users view the crafted content.

What makes this vulnerability particularly dangerous is its potential for escalation. In the Electron desktop application environment, the electronAPI IPC (Inter-Process Communication) bridge is exposed to the renderer process. When an attacker successfully executes JavaScript through the stored XSS, they can leverage this IPC bridge to communicate with the main Electron process, which has access to Node.js APIs and can execute system commands. This transforms a client-side scripting vulnerability into a full Remote Code Execution scenario.

Root Cause

The root cause lies in improper input validation and sanitization within the Mermaid artifact renderer. User-supplied diagram definitions are not adequately sanitized before being processed and rendered, allowing injection of malicious script content. Additionally, the Electron application's security architecture exposes the electronAPI IPC bridge to the renderer context without sufficient isolation, enabling the XSS payload to escalate privileges and execute system-level commands.

Attack Vector

The attack requires local access and high privileges with user interaction. An attacker must craft a malicious Mermaid diagram containing embedded JavaScript. When this content is stored in the application (such as in a chat message or shared document), any user who subsequently views the diagram will execute the malicious script. In the Electron desktop environment, the script can then call methods on the electronAPI object to interact with the main process and execute arbitrary commands on the underlying operating system.

The attack chain involves:

Injecting malicious JavaScript into a Mermaid diagram definition
Storing the malicious content within the LobeChat application
Victim viewing the stored content, triggering XSS execution
Malicious script accessing the electronAPI IPC bridge
Executing system commands through the Electron main process

For detailed technical information about the vulnerability mechanism, refer to the GitHub Security Advisory.

Detection Methods for CVE-2026-23733

Indicators of Compromise

Suspicious Mermaid diagram content containing script tags, event handlers, or JavaScript URIs
Unexpected IPC calls to the electronAPI bridge from renderer processes
Unusual system command execution originating from the LobeChat Electron process
Anomalous network connections or file system access initiated by the application

Detection Strategies

Monitor application logs for malformed or suspicious Mermaid diagram definitions
Implement Content Security Policy (CSP) violation logging to detect inline script execution attempts
Use endpoint detection tools to monitor for suspicious child process spawning from Electron applications
Audit stored user content for potential XSS payloads targeting the Mermaid renderer

Monitoring Recommendations

Enable verbose logging for the Mermaid renderer component to capture processing errors
Monitor Electron main process for unexpected IPC messages or command execution requests
Implement real-time alerting for CSP violations within the application context
Track file system and network activity originating from the LobeChat application process

How to Mitigate CVE-2026-23733

Immediate Actions Required

Upgrade LobeChat to version 2.0.0-next.180 or later immediately
Review stored content for potentially malicious Mermaid diagrams
Consider disabling the Mermaid artifact renderer until patching is complete
Restrict the electronAPI IPC bridge exposure using context isolation and preload scripts

Patch Information

The vulnerability has been patched in LobeChat version 2.0.0-next.180. Organizations should upgrade to this version or later to remediate the vulnerability. The patch addresses the input sanitization issues in the Mermaid artifact renderer, preventing the injection of malicious JavaScript content. For complete patch details, see the GitHub Security Advisory.

Workarounds

Disable the Mermaid artifact renderer feature if not required for business operations
Implement strict Content Security Policy headers to mitigate XSS impact
Enable Electron context isolation to limit renderer process access to Node.js APIs
Use the sandbox option in Electron BrowserWindow configurations to restrict renderer capabilities
Review and restrict the methods exposed through the electronAPI IPC bridge

bash

# Verify LobeChat version to confirm patch status
# Check package.json or application settings for version information
npm list lobe-chat

# Update to patched version
npm update lobe-chat@2.0.0-next.180