CVE-2026-9279: Shell command injection in Logseq

Synopsis

Shell command injection in Logseq allows an attacker with JavaScript execution in the renderer process to run arbitrary shell commands on the host. The CVSS vector (AV:L/PR:N/UI:N) indicates the attacker needs an existing foothold in the renderer, such as through a cross-site scripting flaw or a malicious plugin, but requires no authentication and no victim interaction beyond having Logseq running. Successful exploitation gives the attacker full code execution with the privileges of the Logseq process, including read and write access to the host filesystem and any connected systems in scope. No upstream fix has been published; HarborGuard tracks this advisory and will make a patched-image rebuild available the moment an upstream fix is released.

HarborGuard Coverage

Exploit Conditions

• Network reachabilityNot required

  The attacker needs an existing shell, process, or code-execution foothold on the host (or in the renderer context); no over-the-network path to the vulnerable handler is required.

• AuthenticationNot required

  No credentials or account are required; the IPC handler is reachable from any JavaScript context running inside the Logseq renderer.

• Victim interactionNot required

  No victim action is needed beyond Logseq already running; the attacker's injected JavaScript reaches the IPC handler without any additional user gesture.

• Attack complexityDetail

  Base exploit reliability is high (AC:L), though the CVSS vector notes an attack target precondition (AT:P), meaning specific conditions such as a loaded plugin surface or an XSS entry point must exist before the injection path is reachable.

Blast Radius

• Attacker executes arbitrary shell commands with the full OS-level privileges of the Logseq process, including spawning new processes and writing executable files to disk.
• Any files readable by the Logseq user account, including notes, tokens, SSH keys, and browser cookies stored on the host, are exposed.
• Attacker can write or overwrite files in the user's home directory and any mounted paths accessible to the Logseq process.
• Downstream systems reachable from the host (SC:H/SI:H) are also in scope; the attacker can pivot to network shares, cloud credential files, or other services the host has access to.