Legora before the 2026-08-14 release contains a cross‑site scripting flaw that allows attackers to execute arbitrary JavaScript in a victim’s browser. By embedding a Mermaid diagram prefixed with a gray‑matter front‑matter directive, the front‑matter parser calls eval() before the SVG is sanitized, exposing the browser to script execution. If the victim uses Legora’s Word or Outlook add‑in, the attacker can reach the add‑in’s JavaScript context, potentially reading or exfiltrating session tokens stored in localStorage.

Legora users running the Legora platform prior to the 2026-08-14 update are affected, including the Word and Outlook add‑in versions that persist bearer session tokens in localStorage. No specific build numbers are listed, so all instances before this date should be reviewed.

The CVSS score is 5.1, indicating moderate risk. EPSS data is unavailable, and the vulnerability is not listed in CISA’s KEV catalog, suggesting limited public exploitation data. Attackers can inject malicious Mermaid content through any input channel that renders Mermaid diagrams, such as user‑generated content or files. Because this is a client‑side flaw with no authentication requirement, the threat vector is primarily user‑side injection, but the impact on add‑in contexts raises the potential for credential theft and lateral movement.