SiYuan versions before v3.8.4 contain a path traversal vulnerability in the checkoutRepo endpoint that allows authenticated administrators to write JSON files outside the workspace. Attackers can supply a sessionID parameter containing directory traversal sequences to overwrite arbitrary JSON files in pre-existing kernel-writable directories outside workspace boundaries.
Casky was already ahead
This CVE exploits attack patterns that Casky's 0matched skills already investigate — long before this vulnerability was disclosed. Claude's reasoning model maps these techniques to MITRE ATT&CK, so practitioners who ran these skills have already seen the threat behaviour in their findings.
SiYuan versions before v3.8.4 contain a critical path traversal vulnerability in the checkoutRepo endpoint that enables authenticated administrators to write JSON files outside the intended workspace directory. By injecting directory traversal sequences (such as '../') into the sessionID parameter, attackers can escape sandbox boundaries and overwrite arbitrary JSON files in kernel-writable directories. This vulnerability matters because it breaks the fundamental assumption that administrative actions remain confined to workspace boundaries—attackers with admin credentials can now corrupt system configurations, inject malicious settings, or compromise data integrity across the host system. Organizations running vulnerable SiYuan instances with untrusted administrators or compromised admin accounts face significant risk of lateral movement and system compromise.
While this CVE lacks direct MITRE ATT&CK technique mappings in the vulnerability record, Casky's skill-based detection framework would identify this attack through techniques like T1083 (File and Directory Discovery) and T1565 (Data Manipulation) by analyzing anomalous file write operations outside workspace boundaries. Practitioners using Casky would observe detection signals when the platform's Claude-powered analysis identifies: (1) sessionID parameters containing traversal sequences like '../' or '..\', (2) JSON file writes to unexpected system paths outside the workspace, (3) authentication context (admin accounts) combined with path traversal attempts, and (4) modifications to kernel-writable directories typically outside application scope. The extended reasoning capability would correlate these signals into a coherent attack narrative, helping security teams distinguish between legitimate administrative activity and exploitation attempts targeting this specific vulnerability class.
Composite risk scoring from EPSS, CISA KEV, Shodan, and GreyNoise — 21 security APIs correlated into a single Casky Risk Score. Coming in Casky Pro. Join early access →
Casky has 0 skills that investigate the attack patterns behind CVE-2026-100637. Run one and get CVSS-scored findings in 3 minutes.
Run the skill that detects this →© 2026 Casky.AI, Inc. · AI Security Investigation