CVE-2026-11422: Markdown Preview Enhanced Code Injection via WaveDrom Rendering

Technical summary

CVE-2026-11422 is a code injection vulnerability (CWE-95: Improper Neutralization of Directives in Dynamically Evaluated Code) in Markdown Preview Enhanced's WaveDrom rendering pipeline. The extension passes unsanitized content from WaveDrom fenced code blocks directly to window.eval() within the VS Code webview context. An attacker can embed malicious JavaScript in a wavedrom code block within a Markdown document. When the document is previewed, the JavaScript executes in the webview sandbox. By exploiting the extension's message-passing mechanism, the attacker can escalate from code execution in the webview to arbitrary file operations on the local filesystem, bypassing the webview's isolation boundary.

Business impact

For organizations where developers use Markdown Preview Enhanced to review technical documentation, design diagrams, or specifications, this vulnerability creates a significant supply-chain risk. A single malicious Markdown file—shared via email, repository, documentation system, or wiki—can compromise a developer's workstation. Once code execution is achieved, an attacker can exfiltrate source code, credentials stored in environment files, SSH keys, or internal documentation. The ability to write files enables persistence mechanisms, lateral movement tooling, or outright malware installation. The user-interaction requirement (opening a preview) is a minor friction point in threat scenarios involving social engineering or trojanized documentation repositories.

Affected systems

Markdown Preview Enhanced version 0.8.x in combination with crossnote engine 0.9.28 is vulnerable. VS Code is the primary affected platform, though any host running the extension that processes untrusted Markdown files is at risk. Developers, technical writers, security researchers, and documentation teams who routinely open Markdown files from external or less-trusted sources face the highest exposure.

Exploitability

Exploitation requires user interaction—the victim must open a Markdown document in VS Code with Markdown Preview Enhanced active. The attack vector is local (requires file system access to place or open the file), but delivery is straightforward: the attacker can embed the payload in a GitHub repository, attach it to an email, or place it on a shared drive. No special privileges, authentication, or knowledge of the target system is required. The barrier to weaponization is low; creating a proof-of-concept requires only knowledge of WaveDrom syntax and basic JavaScript. The lack of KEV (Known Exploited Vulnerabilities) status as of the publication date suggests active exploitation in the wild has not been formally documented, but the simplicity of the attack makes exploitation probable.

Remediation

Upgrade Markdown Preview Enhanced to a patched version that sanitizes WaveDrom block content before passing it to any evaluation function. Additionally, ensure the crossnote engine dependency is updated to a version that does not contain this vulnerability. Verify the specific patch versions against the official Markdown Preview Enhanced repository or the VS Code extension marketplace. As an interim measure, disable the Markdown Preview Enhanced extension or restrict preview of untrusted Markdown files until patching is confirmed.

Patch guidance

Check the Markdown Preview Enhanced changelog and VS Code extension marketplace for version releases published after June 17, 2026 (the modification date of this CVE). The vendor should have released a patched version of 0.8.x or later that removes unsanitized eval() calls and properly sanitizes WaveDrom input. Update via the VS Code Extensions panel or verify the specific patched version number against the official vendor advisory before deployment. Test in a non-production environment to confirm the patch does not break legitimate WaveDrom rendering functionality.

Detection guidance

Monitor VS Code extension logs for errors or unusual activity following Markdown preview operations. Endpoint detection and response (EDR) tools should flag suspicious file write operations originating from VS Code webview processes. Network monitoring can identify unexpected outbound connections from the extension. File integrity monitoring on user home directories and credential storage locations (.ssh, .aws, etc.) may detect unauthorized modifications. Consider scanning Markdown repositories and shared documentation systems for suspicious WaveDrom syntax patterns (e.g., WaveDrom blocks containing JavaScript event handlers or eval() calls). On a preventive basis, restrict installation of Markdown Preview Enhanced in managed development environments to verified, patched versions.

Why prioritize this

This vulnerability merits urgent attention due to its HIGH CVSS score (7.1), direct path to code execution and file system access, and low barrier to exploitation. While it requires user interaction, developers and technical teams often work with Markdown files as part of normal workflow, making social engineering effective. The ability to write arbitrary files creates persistence and lateral movement opportunities. The lack of privilege requirement means even non-administrator users are fully exploitable. Organizations should prioritize patching before the vulnerability is widely exploited or incorporated into automated attack toolkits.

Risk score, explained

CVSS 7.1 (HIGH) reflects: (1) Local attack vector—the file must be accessed locally, limiting mass-exploitation but not significantly reducing risk in developer environments; (2) Low attack complexity—no special conditions or system knowledge required; (3) No privilege requirement—any user can trigger the vulnerability; (4) User interaction required—reduces the score but remains a minor barrier in social engineering scenarios; (5) Confidentiality and integrity impact—the attacker can read (C:H) and write (I:H) arbitrary files; (6) No availability impact—the attack does not directly disrupt system functioning. The score appropriately captures the severity: this is a dangerous vulnerability for any organization using the extension, but it is not a full system compromise or network-wide threat vector.

Frequently asked questions