Siyuan

Directory traversal in SiYuan personal knowledge management system allows authenticated attackers to read arbitrary workspace files via double URL encoding bypass. The vulnerability stems from an incomplete fix for CVE-2026-30869 that added only denylist validation without removing a redundant url.PathUnescape() call in serveExport(). Attackers can use %252e%252e encoding to access sensitive files including the complete SQLite database (siyuan.db), kernel logs, and all user documents. EPSS data not available for this recent CVE; publicly available exploit code exists (GitHub commit demonstrates exploitation technique).

SiYuan is an open-source personal knowledge management system. In versions 3.6.3 and prior, the /api/av/removeUnusedAttributeView endpoint constructs a filesystem path using the user-controlled id parameter without validation or path boundary enforcement. An attacker can inject path traversal sequences such as ../ into the id value to escape the intended directory and delete arbitrary .json files on the server, including global configuration files and workspace metadata. This issue has been fixed in version 3.6.4.

SiYuan 3.6.1 through 3.6.3 allows arbitrary code execution when users view malicious bazaar packages in the marketplace UI. The vulnerability stems from an incomplete XSS fix (for CVE-2026-33066) that enabled an HTML sanitizer but failed to block iframe tags with srcdoc attributes containing embedded scripts. A malicious package author can inject JavaScript that executes in the Electron process with full application privileges, compromising the user's machine. The issue is confirmed fixed in version 3.6.4 and no public exploitation has been reported at time of analysis.

Cross-site scripting (XSS) in SiYuan personal knowledge management system versions 3.6.0-3.6.1 allows remote attackers to execute arbitrary JavaScript via the unauthenticated /api/icon/getDynamicIcon endpoint by bypassing SVG sanitization using XML namespace-prefixed element names. The vulnerability exploits a parser inconsistency where Go's HTML5 parser treats 'x:script' as a distinct tag while browsers' XML parsers resolve the namespace prefix to execute the script. Publicly available exploit code exists (GitHub issue #17246 documents the bypass technique), though no CISA KEV listing indicates mass exploitation campaigns at time of analysis.

Stored cross-site scripting (XSS) in SiYuan personal knowledge management system versions prior to 3.6.2 escalates to remote code execution in the Electron desktop client. Attackers craft malicious .sy.zip import files containing HTML entities mixed with raw special characters that bypass server-side attribute escaping, injecting event handlers into imported notes. When victims open the compromised note in the Electron client, injected JavaScript executes with full Node/Electron API access, enabling arbitrary code execution. CVSS 8.6 (High) with local attack vector requiring user interaction; no public exploit identified at time of analysis.

Unauthenticated information disclosure in SiYuan personal knowledge management system versions before 3.6.2 allows remote attackers to retrieve confidential content from password-protected documents via the publish service's bookmark API endpoint. The vulnerability bypasses document-level access controls by treating nil authentication contexts as authorized during bookmark filtering, exposing any bookmarked blocks without password verification. CVSS 7.5 (High) with network-based exploitation requiring no privileges or user interaction; no public exploit identified at time of analysis, though the security advisory provides detailed technical disclosure.

Stored cross-site scripting (XSS) in SiYuan personal knowledge management system escalates to arbitrary operating system command execution on desktop clients. Authenticated attackers with low privileges can inject malicious URLs into Attribute View asset fields that execute JavaScript when victims view Gallery or Kanban layouts with "Cover From -> Asset Field" enabled. The Electron desktop client's configuration (nodeIntegration enabled, contextIsolation disabled) allows the XSS payload to break sandbox boundaries and execute arbitrary commands under the victim's OS account. CVSS 9.0 (Critical) with network attack vector, low complexity, and cross-scope impact. Vendor-released patch: version 3.6.2. No public exploit identified at time of analysis, though technical details are disclosed in GitHub advisory GHSA-rx4h-526q-4458.

High severity vulnerability in SiYuan Note. # The `/api/network/forwardProxy` endpoint allows authenticated users to make arbitrary HTTP requests from the server. The endpoint accepts a user-controlled URL and makes HTTP requests to it, returning the full response body and headers. There is no URL validation to prevent requests to internal networks, localhost, or cloud metadata services.

SiYuan's SVG sanitizer fails to properly filter malicious href attributes when whitespace characters are inserted into javascript: URLs, allowing reflected cross-site scripting on the unauthenticated /api/icon/getDynamicIcon endpoint. Public exploit code exists for this vulnerability, which bypasses the previous fix for CVE-2026-29183. Attackers can inject executable JavaScript to target unauthenticated users of SiYuan versions prior to 3.5.10.

Reflected XSS in SiYuan's /api/icon/getDynamicIcon endpoint allows unauthenticated attackers to inject malicious JavaScript through SVG animation elements that bypass the sanitizer's static filters. The vulnerability exists because the SVG sanitizer blocks script tags and event handlers but fails to restrict <animate> and <set> elements, which can dynamically modify attributes at runtime to execute code. Public exploit code exists and patches are not yet available for affected versions prior to 3.5.10.

SiYuan prior to 3.5.10 has a path traversal vulnerability enabling arbitrary file access through crafted API requests.

SiYuan Note prior to version 3.5.10 contains an insufficient authorization flaw in the /api/block/appendHeadingChildren endpoint that allows authenticated users with read-only (RoleReader) privileges to modify notebook content by appending blocks to documents. The vulnerability exists because the endpoint applies only basic authentication checks instead of enforcing stricter administrative or read-only restrictions. Affected users should upgrade to version 3.5.10 or later, as no workaround is currently available and exploitation requires only network access and valid read-only credentials.

Reflected XSS in SiYuan knowledge management before 3.5.9.

SQL injection in SiYuan prior to version 3.6.0 allows any authenticated user, including those with read-only access, to execute arbitrary database queries through the /api/query/sql endpoint due to insufficient authorization checks. Public exploit code exists for this vulnerability, enabling attackers to extract sensitive data or modify the knowledge base contents. No patch is currently available for affected versions.

Unauthenticated attackers can read sensitive configuration files from SiYuan knowledge management systems prior to version 3.5.5 by exploiting case-sensitivity bypass in file access controls on Windows and other case-insensitive filesystems. The /api/file/getFile endpoint fails to properly validate mixed-case path traversal attempts, allowing unauthorized access to protected data. Public exploit code exists for this vulnerability, and no patch is currently available.

Stored XSS in Lute's Markdown rendering engine (versions 1.7.6 and earlier) allows authenticated attackers to inject malicious JavaScript into notes that executes when other users view the rendered content. SiYuan and other applications using vulnerable Lute versions are affected, with public exploit code available. A patch is available and should be applied to prevent session hijacking and credential theft.

SiYuan knowledge management system prior to 3.5.5 has a path traversal in /api/file/copyFile allowing arbitrary file operations on the server.

SiYuan personal knowledge management system prior to 3.5.4 has a stored XSS vulnerability (CVSS 9.6) that allows code execution through crafted knowledge base entries.

SiYuan knowledge management system versions before 3.5.4 allow authenticated users to copy arbitrary files from the server filesystem into the application workspace due to insufficient path validation in the /api/file/globalCopyFiles endpoint. An attacker with valid credentials can exploit this path traversal vulnerability to read sensitive files and escalate privileges within the application. Public exploit code exists for this medium-severity vulnerability, though a patch is available.

Remote attackers can read arbitrary files from SiYuan servers (versions prior to 3.5.4) by exploiting server-side HTML rendering in the markdown feature. The path traversal vulnerability (CWE-22) requires no authentication and has low attack complexity, making it trivially exploitable. A public exploit exists and EPSS scoring indicates 9% exploitation probability (25th percentile), suggesting limited but active reconnaissance. Vendor patch available in version 3.5.4.

Reflected XSS in SiYuan's /api/icon/getDynamicIcon endpoint allows attackers to inject malicious JavaScript through unescaped SVG content in dynamically generated icon images. An unauthenticated attacker can craft a malicious link that, when clicked by a victim, executes arbitrary scripts in the context of the SiYuan application. Public exploit code exists for versions prior to 3.5.4, which contains the necessary patches.

SiYuan prior to version 3.5.4-dev2 fails to sanitize SVG file uploads, allowing authenticated attackers to embed malicious JavaScript that executes when other users view the files. Public exploit code exists for this stored XSS vulnerability, which can compromise user sessions and access sensitive knowledge management data. The vulnerability affects self-hosted instances where users can upload SVG content from untrusted sources.

SiYuan is self-hosted, open source personal knowledge management software. Rated high severity (CVSS 8.7), this vulnerability is remotely exploitable, no authentication required, low attack complexity. Public exploit code available.