108
CVEs
30
Critical
39
High
0
KEV
36
PoC
4
Unpatched C/H
94.4%
Patch Rate
0.3%
Avg EPSS
Severity Breakdown
CRITICAL
30
HIGH
39
MEDIUM
38
LOW
1
Monthly CVE Trend
Affected Products (4)
Top Risky CVEs
| CVE | Summary | Severity | CVSS | EPSS | Priority | Signals |
|---|---|---|---|---|---|---|
| CVE-2026-69084 | SQL injection in SiYuan personal knowledge management (versions <= v3.7.2) lets attackers pass arbitrary SQL statements verbatim to the primary read-write siyuan.db handle via the /api/search/searchEmbedBlock endpoint. Because the endpoint is gated only by CheckAuth, it is reachable with the low-privilege publish RoleReader token - and by fully anonymous users when publish authentication is disabled - and the driver's support for stacked statements allows both reading and modifying content across all opened cleartext notebooks. No public exploit identified at time of analysis, but the flaw was reported by VulnCheck and carries a critical 9.9 CVSS 4.0 score; it is fixed in v3.7.3. | CRITICAL | 9.9 | 0.3% | 70 |
PoC
|
| CVE-2026-54069 | Origin-validation bypass in SiYuan Note (open-source personal knowledge management) before 3.7.0 lets any installed Chrome/Chromium browser extension obtain RoleAdministrator access to the local kernel HTTP server at 127.0.0.1:6806. Because the kernel unconditionally trusts all chrome-extension:// origins and desktop installs ship with an empty AccessAuthCode by default, a malicious or supply-chain-compromised extension can issue fully authenticated admin API calls with no further authentication, enabling data exfiltration, stored XSS injection, and configuration tampering. Publicly available exploit code exists; there is no public exploit identified as actively exploited. | CRITICAL | 9.2 | 0.6% | 67 |
PoC
|
| CVE-2026-54066 | Arbitrary file disclosure in SiYuan personal knowledge management system before 3.7.0 lets an unauthenticated remote attacker read any file inside the workspace directory through its publish-mode HTTP endpoint (default port 6808). The flaw is an incomplete fix for CVE-2026-41894: the earlier patch sanitized the /export/ route but left the identical double-URL-encoding path traversal in the /assets/*path route, exposing conf/conf.json (AccessAuthCode SHA256 hash, API token, sync keys), the temp SQLite databases, and siyuan.log. Publicly available exploit code exists; no public active exploitation has been confirmed. | HIGH | 7.5 | 1.9% | 59 |
PoC
|
| CVE-2026-66012 | Unauthenticated remote administrator takeover in SiYuan before v3.7.2 via missing authorization on the /mcp kernel endpoint. When the Publish server is in anonymous mode, a remote attacker can invoke 31 MCP tools—including full file-system read, write, and delete—to exfiltrate secrets from conf/conf.json and plant a malicious plugin that runs with elevated desktop privileges. While no active exploitation has been confirmed by CISA KEV, detailed exploit information is publicly available, and the vulnerability carries a maximum CVSS severity. | CRITICAL | 10.0 | 0.4% | 50 |
|
| CVE-2026-50551 | Stored cross-site scripting in SiYuan's Attribute View (database) asset cell renderer escalates to remote code execution on the Electron desktop client for all versions prior to 3.7.0. An attacker who can persist crafted content into a database asset cell has malicious JavaScript executed in the privileged Electron context, breaking out to the host operating system; the issue carries a 9.9 CVSS and is fixed in 3.7.0. No public exploit identified at time of analysis, and the flaw is not listed in CISA KEV. | CRITICAL | 9.9 | 0.4% | 50 |
|
| CVE-2026-69083 | SQL injection in SiYuan's fullTextSearchAssetContent endpoint lets unauthenticated attackers and holders of publish RoleReader tokens execute arbitrary SQL against the read-write asset-content database in all versions before v3.7.3. Because the endpoint is reachable without authentication and the database is writable, attackers can read, alter, or delete data across notebooks - not merely exfiltrate it. No public exploit identified at time of analysis, but the flaw was reported by VulnCheck and carries a CVSS 4.0 base score of 9.9. | CRITICAL | 9.9 | 0.4% | 50 |
PoC
|
| CVE-2026-54067 | Stored cross-site scripting in SiYuan personal knowledge management before 3.7.0 lets an attacker with write access to a synced workspace plant a CSS snippet whose '</style>' sequence breaks out of the surrounding <style> tag when renderSnippet() interpolates it via insertAdjacentHTML, running arbitrary JavaScript in the renderer. On Electron desktop builds the renderer runs with nodeIntegration:true, so the injected handler can reach require('child_process') and escalate the XSS to full host remote code execution. The payload syncs through the workspace repository and fires automatically on every device that pulls, and it bypasses the user's enabledJS=off setting; no public exploit or CISA KEV listing is identified at time of analysis. | CRITICAL | 9.9 | 0.3% | 50 |
PoC
|
| CVE-2026-54158 | Stored cross-site scripting in SiYuan's attribute-view (database) cell renderer, affecting all versions prior to 3.7.0, lets a low-privileged attacker with write access to a synced workspace plant a persistent payload that runs arbitrary JavaScript when a victim opens the block-attribute panel. The genAVValueHTML function interpolates cell content raw in its text, url, phone, and mAsset branches and the Go kernel never escapes the value on input, so the malicious row survives sync byte-for-byte and fires on every device that renders it; on the Electron desktop client (nodeIntegration:true) the XSS escalates to host remote code execution via require('child_process'). CVSS is rated 9.9; there is no public exploit identified at time of analysis and the issue is not listed in CISA KEV. | CRITICAL | 9.9 | 0.3% | 50 |
PoC
|
| CVE-2026-69085 | SQL injection in SiYuan personal knowledge-management software (all versions before v3.7.3) lets attackers inject arbitrary SQL through the unescaped keyword parameter of the /api/filetree/searchDocs endpoint. Because the query runs on a read-write SQLite handle that permits stacked statements, an attacker can both read and tamper with content across every cleartext notebook, and the endpoint is reachable unauthenticated when publish mode runs with Publish.Auth.Enable=false. Reported by VulnCheck with a vendor patch available; no public exploit identified at time of analysis, and the CVSS 4.0 score is 9.9. | CRITICAL | 9.9 | 0.2% | 50 |
|
| CVE-2026-72811 | SQL injection in SiYuan's backlink/mention search (kernel/model/backlink.go) lets unauthenticated attackers execute arbitrary SQL against the self-hosted note-taking application through v3.7.2. Because the search routine escapes only double quotes and leaves single quotes intact, a crafted keyword or malicious stored block metadata breaks out of the SQL string literal on the primary read-write siyuan.db handle, enabling cross-notebook read and write. VulnCheck reports the flaw (CVSS 4.0 base 9.9); no public exploit identified at time of analysis, and it is fixed in v3.7.4. | CRITICAL | 9.9 | 0.2% | 50 |
PoC
|
| CVE-2026-74800 | Stored cross-site scripting in SiYuan before v3.7.4 lets an authenticated user upload an HTML file through the asset-serving endpoint, which the server returns without Content-Disposition or X-Content-Type-Options headers, so the browser renders it inline instead of downloading it. When the workspace owner clicks the malicious asset link, the injected script runs in the app's origin with full kernel API access, enabling account and data compromise. Reported by VulnCheck with a vendor patch available; no public exploit has been identified at time of analysis. | CRITICAL | 9.4 | 0.3% | 47 |
|
| CVE-2026-73056 | Unthrottled API-token brute force in SiYuan kernel before 3.7.4 lets an unauthenticated remote attacker repeatedly guess the Conf.Api.Token because the CheckAuth() middleware bypasses the app's CAPTCHA/lockout controls (NeedCaptcha/WrongAuthCount). Reported by VulnCheck and fixed in 3.7.4, a successful guess yields full RoleAdministrator access, permitting arbitrary file operations and SQL queries. There is no public exploit identified at time of analysis and it is not in CISA KEV, though the CVSS 4.0 base score is 9.3. | CRITICAL | 9.3 | 0.4% | 47 |
|
| CVE-2026-74798 | Arbitrary file read and deletion in SiYuan kernel before v3.7.4 arises from an unvalidated id parameter in the database_clean MCP tool, letting an authenticated MCP client traverse the filesystem to copy any process-readable file into the history directory and then delete the original. This is the MCP-facing twin of a flaw previously fixed on the HTTP API path (GHSA-7hm9-v7vf-7g4w), meaning a known-bad code pattern survived in a second caller. Reported by VulnCheck with a vendor patch available; no public exploit identified at time of analysis and not listed in CISA KEV. | CRITICAL | 9.3 | 0.4% | 47 |
|
| CVE-2026-82653 | Stored cross-site scripting in SiYuan before v3.8.1 lets a low-privileged attacker persist HTML/script payloads in bazaar package names and notebook names, which the confirmDialog() function interpolates unescaped into innerHTML. The script executes in a victim's browser session when they uninstall a malicious package or unlock an encrypted notebook, enabling session-context takeover of the note-taking application. No public exploit identified at time of analysis and no active exploitation is recorded. | CRITICAL | 9.3 | 0.2% | 47 |
|
| CVE-2026-82654 | Stored cross-site scripting in SiYuan before v3.8.1 lets an authenticated user plant HTML/script payloads in a block's name, alias, or memo field, which then execute in another user's session when that user views any document that references or displays the poisoned block via hint, backlink, or breadcrumb rendering. Because the malicious markup runs in the victim's authenticated context, an attacker can hijack sessions and act as the victim. No public exploit identified at time of analysis, and the flaw is not listed in CISA KEV. | CRITICAL | 9.3 | 0.2% | 47 |
|