Logseq vulnerabilities

Logseq exposes an IPC handler that allows the renderer process to execute shell commands. While an allowlist restricts the command name, the argument string is concatenated with the command and passed to child_process.spawn with the shell: true option, allowing shell metacharacters in the arguments to bypass the allowlist. An attacker with JavaScript execution in the renderer can execute arbitrary shell commands with the privileges of the Logseq process, leading to remote code execution on the host.

Logseq is vulnerable to a sandbox escape flaw where plugins running in sandboxed iframes can inject arbitrary HTML attributes (e.g., event handlers) into their container element in the host DOM. Due to a disabled CSP, a malicious plugin can execute arbitrary JavaScript in the privileged host context, potentially gaining unauthorized access to filesystem APIs. Only version v0.10.15 was tested and confirmed vulnerable; status of other versions is unknown.

Logseq is vulnerable to stored XSS. A malicious plugin can include JavaScript in the "name" field of its package.json, rendered via innerHTML without sanitization, allowing arbitrary code execution in the host context.

The Electron preload script in Logseq exposes an API method that allows the renderer process to invoke IPC handlers without proper path validation. An attacker with JavaScript execution in the renderer (e.g. via XSS or a malicious plugin) can read, write, or delete arbitrary files on the user's system. The vulnerability was confirmed in version v0.10.15, status of other versions is unknown.