IntelliJ IDEA CVE-2026-64810

Exploitation requires that the targeted user is running IntelliJ IDEA prior to version 2026.2 and views an IDE notification containing attacker-influenced content (UI:R per CVSS). The attacker must have a delivery mechanism to inject HTML into the notification pipeline - viable vectors include a malicious or compromised plugin, a crafted project file whose metadata is rendered in notifications, or a remote server response surfaced by an IDE feature (e.g., version check, plugin repository). The CVSS PR:N indicates no authentication to a remote system is required on the attacker's side to deliver the payload. Exploitation is strictly limited to confidentiality impact in the form of silent activity tracking; the attacker cannot modify files, execute code, or disrupt service (I:N/A:N). Scope is unchanged (S:U), meaning impact is confined to the IDE process and does not extend to the underlying OS or other applications.

The CVSS 4.3 Medium score is consistent with a narrowly scoped, confidentiality-only impact: limited data exposure (C:L), no integrity or availability impact (I:N/A:N), network attack vector (AV:N), low complexity (AC:L), no authentication required (PR:N), but requiring user interaction (UI:R). EPSS data was not supplied in the available intelligence, so exploitation probability cannot be quantified. KEV status is absent, meaning CISA has not confirmed active exploitation. No POC exploit code has been identified at time of analysis. The primary real-world risk is covert developer activity tracking - a meaningful privacy and potential corporate espionage concern rather than a system compromise vector. In enterprise environments where developer workstation activity is sensitive (e.g., defense contractors, financial institutions), the silent tracking capability elevates practical impact beyond the raw CVSS score suggests. The attack's dependence on UI:R and on the attacker's ability to inject content into the notification pipeline are meaningful limiting factors.

An attacker distributes a malicious IntelliJ IDEA plugin or crafts a repository whose project metadata surfaces attacker-controlled content as an IDE notification. When the targeted developer views the notification in IntelliJ IDEA before 2026.2, an injected HTML image tag silently loads a remote resource from an attacker-controlled server, logging the request timestamp and IP address to track developer activity. No public exploit code has been identified at time of analysis, but the technique is well-understood and consistent with standard HTML injection patterns.