On this page (7 sections)
A critical stored cross-site scripting (XSS) flaw, tracked as CVE-2026-130679, has been published for SiYuan, the open source note-taking and knowledge management application. According to the GitHub Advisory Database entry, versions before 3.8.10 do not escape an image source value in the Attribute View feature, which lets stored script run when a user opens an affected note or row attribute panel.
The notable part is the impact in the desktop app: the advisory says the injected script can execute operating system commands through Node.js. If you or your team run SiYuan, updating to 3.8.10 or later (the version named in the advisory description) is the sensible first step.
What happened
On November 18, 2026 the vulnerability was published through the National Vulnerability Database and mirrored in the GitHub Advisory Database. The description names a function, genAVValueHTML(), which builds the HTML shown for Attribute View values.
According to the advisory, an image source value is placed into that HTML without escaping. An attacker who is able to store Attribute View asset values can store a crafted image URL. When a victim opens the bound note or the row attribute panel, the script executes.
Technical details in plain terms
Stored XSS happens when an application saves attacker-supplied text and later writes it into a page as HTML without encoding special characters. Here, the advisory says the image src attribute is unescaped, and that a stored image URL containing a double quote and an onerror handler is what triggers script execution.
The reason this is rated critical is the desktop app. The advisory states that the injected script executes OS commands via Node.js in the desktop app. Commands would run with the permissions of the user who opened the content, which is the usual consequence of code execution on a workstation (general reasoning, not a detail from the advisory).
The CVSS v4 metrics listed in the advisory are: network attack vector, low attack complexity, no attack requirements, low privileges required, passive user interaction, and high impact to confidentiality, integrity and availability for both the vulnerable and subsequent systems.
No exploit details beyond the advisory description are available in the sources, and none are repeated here.
Who is exposed
The advisory states two conditions: an attacker must be able to store Attribute View asset values, and a victim must open the bound note or row attribute panel. It describes OS command execution in the desktop app. Beyond that, the sources do not say which deployments are affected or whether anyone was targeted.
Our own reasoning about plausible exposure, not findings from the advisory:
- Anyone running the SiYuan desktop app on a version before 3.8.10 is in the case the advisory describes.
- Workspaces where more than one person can add or edit content, or where you open notes or databases that others created, offer more ways for a crafted value to be stored.
- Content imported from untrusted sources is worth treating with caution, since it could carry values you did not write.
- Other deployment types (for example a server instance) are not described in the advisory, so their impact is unknown.
How to check your exposure
Find your SiYuan version first. In the desktop app, open the About or settings page and read the version number. Treat anything below 3.8.10 as affected, based on the advisory description.
Then review who can write to your workspaces: list collaborators, shared notebooks, and recent imports from outside your organization. Pay attention to Attribute View image or asset values you did not create, especially values containing quote characters or event-handler text such as onerror.
What to do now
- Update SiYuan to version 3.8.10 or later on every device where it is installed, per the advisory description.
- Fully quit and restart the app after updating so the old code is no longer running.
- Until you have updated, avoid opening notes or Attribute View rows from untrusted sources or recently imported data.
- Review shared workspaces and remove edit access for people who do not need it.
- Inspect Attribute View asset or image values for unexpected content and remove anything you cannot explain.
- If you opened untrusted content on a vulnerable version, treat the computer as possibly exposed: change passwords and tokens used on it and look for unfamiliar processes or scheduled tasks.
- Check the upstream fix commit and project security advisory for confirmed affected and patched version details.
Detection and monitoring afterwards
The following is general endpoint guidance, not something taken from the advisory. Because the flaw runs inside the app on a user's machine, outside-in website checks are unlikely to reveal it. Endpoint tools are a better fit: watch for the SiYuan process starting shells or other unexpected child processes, and for new startup entries or unfamiliar outbound connections.
If you also expose any web service to the internet, keep its certificate and uptime under watch. A certificate and uptime monitor and the free SSL checker help you notice changes on a public endpoint, though they do not detect this flaw.
More broadly, keep a software inventory and a routine for applying critical fixes. Our guide to small business website security covers that.
Timeline
| Date | Event |
|---|---|
| 2026-11-18 | CVE-2026-130679 published by the National Vulnerability Database |
| 2026-11-18 | Advisory published and last updated in the GitHub Advisory Database (marked Unreviewed) |
Sources
This article is based on the public sources below. Details can change as vendors and researchers publish more; check the original advisories before acting on version numbers.
