Description

SiYuan before v3.7.4 contains a cross-site scripting vulnerability in the file upload validation flow that fails to escape filenames before inserting them into HTML via insertAdjacentHTML. Attackers can craft a malicious filename containing script payloads that execute with full OS command access when a user drags, drops, or pastes the file into the editor.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "8.6",
  "pubDate": "2026-08-18T12:19:30.370Z",
  "pubdate": "2026-08-18T12:19:30.370Z",
  "executiveSummary": "An unauthenticated cross-site scripting (XSS) vulnerability exists within SiYuan prior to v3.7.4, specifically residing in the file upload validation flow. The security flaw stems from the application's failure to properly sanitize and escape filenames before dynamically inserting them into the Document Object Model (DOM) via the insertAdjacentHTML method.\nThe primary impact of this vulnerability is the execution of arbitrary scripts within the context of the affected user's session. Due to the high privilege level and potential integrations of the application, successful exploitation allows an attacker to achieve full operating system command access upon the execution of the payload.\nThe affected product is SiYuan in versions prior to v3.7.4. The risk implications are severe, as attackers can weaponize local file interactions to escalate privileges and compromise the host system.\nAttacker capabilities include the creation and delivery of specially crafted malicious files designed with script payloads embedded directly within the filename string. Exploitation requirements mandate that a user interacts with the application by dragging, dropping, or pasting the malicious file directly into the editor interface, triggering the insecure DOM insertion routine.",
  "technicalDetails": "The root cause of the vulnerability resides in the file upload and rendering logic of SiYuan prior to v3.7.4. When files are processed via drag-and-drop or paste actions within the editor component, the application extracts the filename metadata and incorporates it directly into the user interface.\nThe vulnerable component is the file upload validation and rendering flow, which improperly handles untrusted filename inputs by utilizing the insertAdjacentHTML function without prior context-aware output encoding or sanitization. Because the filename is treated as raw HTML rather than a literal string, embedded script tags or event handlers within the filename are parsed and executed by the rendering engine.\nThe attack flow proceeds as follows: First, an attacker crafts a malicious file utilizing a specially engineered payload string as the filename, embedding arbitrary JavaScript commands. Second, the attacker induces a victim to drag, drop, or paste this malicious file into the SiYuan editor interface. Third, the application reads the file metadata, including the malicious filename, and invokes the vulnerable insertion routine via insertAdjacentHTML. Fourth, the browser parses the injected string, executing the embedded script payload within the security context of the application session.\nRegarding environmental and access constraints, the vulnerability can be triggered locally through user interaction without requiring prior authentication. However, the post-exploitation impact is critical, as execution of the payload within the desktop application context yields full operating system command access, allowing the attacker to interact with the underlying host system with the privileges of the running application process."
}