JunoClaw is an agentic AI platform built on Juno Network. Prior to 0.x.y-security-1, plugin-shell's run_command wrapped every agent-supplied command in 'sh -c' / 'cmd /C' and passed the full argument string to the shell's parser, allowing shell metacharacters in agent-supplied arguments to be interpreted as command syntax. This vulnerability is fixed in 0.x.y-security-1.

All versions of the package keep-module-latest are vulnerable to Command Injection due to missing input sanitization or other checks and sandboxes being employed to the installModule function. **Note:** To execute the code snippet and potentially exploit the vulnerability, the attacker needs to have the ability to run Node.js code within the target environment. This typically requires some level of access to the system or application hosting the Node.js environment.

Anthropic Claude Code CLI and Claude Agent SDK contain an OS command injection vulnerability in the command lookup helper and deep-link terminal launcher that allows local attackers to execute arbitrary commands by manipulating the TERMINAL environment variable. Attackers can inject shell metacharacters into the TERMINAL variable which are interpreted by /bin/sh when the command lookup helper constructs and executes shell commands with shell=true. The vulnerability can be triggered during normal CLI execution as well as via the deep-link handler path, resulting in arbitrary command execution with the privileges of the user running the CLI.

OpenClaw versions prior to 2026.2.14 contain an arbitrary file read vulnerability in the exec-approvals allowlist validation that checks pre-expansion argv tokens but executes using real shell expansion. Attackers with authorization or through prompt-injection attacks can exploit safe binaries like head, tail, or grep with glob patterns or environment variables to disclose files readable by the gateway or node process when host execution is enabled in allowlist mode.

systeminformation is a System and OS information library for node.js. In versions prior to 5.30.8, a command injection vulnerability in the `wifiNetworks()` function allows an attacker to execute arbitrary OS commands via an unsanitized network interface parameter in the retry code path. In `lib/wifi.js`, the `wifiNetworks()` function sanitizes the `iface` parameter on the initial call (line 437). However, when the initial scan returns empty results, a `setTimeout` retry (lines 440-441) calls `getWifiNetworkListIw(iface)` with the **original unsanitized** `iface` value, which is passed directly to `execSync('iwlist ${iface} scan')`. Any application passing user-controlled input to `si.wifiNetworks()` is vulnerable to arbitrary command execution with the privileges of the Node.js process. Version 5.30.8 fixes the issue.

virtualenv before 20.26.6 allows command injection through the activation scripts for a virtual environment. Magic template strings are not quoted correctly when replacing. NOTE: this is not the same as CVE-2024-9287.

OpenClaw is a personal AI assistant. Prior to 2026.1.20, an unauthenticated local client could use the Gateway WebSocket API to write config via config.apply and set unsafe cliPath values that were later used for command discovery, enabling command injection as the gateway user. This vulnerability is fixed in 2026.1.20.

Figma Desktop for Windows version 125.6.5 contains a command injection vulnerability in the local plugin loader. An attacker can execute arbitrary OS commands by setting a crafted build field in the plugin's manifest.json. This field is passed to child_process.exec without validation, leading to possible RCE. NOTE: this is disputed by the Supplier because the behavior only allows a local user to attack himself via a local plugin. The local build procedure, which is essential to the attack, is not executed for plugins shared to the Figma Community.

A command injection vulnerability has been reported to affect HybridDesk Station. If an attacker gains local network access, they can then exploit the vulnerability to execute arbitrary commands. We have already fixed the vulnerability in the following version: HybridDesk Station 4.2.18 and later

The issue occurs because tagName user input is formatted inside the exec function is executed without any checks.

An unauthenticated local attacker can inject a command that is subsequently executed as root, leading to a privilege escalation.

OS command injection vulnerabilities in GE HealthCare ultrasound devices

There is an OS command injection vulnerability in Crestron AM-300 firmware version 1.4499.00018 which may enable a user of a limited-access SSH session to escalate their privileges to root-level access.

IBM AIX 7.1, 7.2, 7.3, and VIOS 3.1 could allow a non-privileged local user to exploit a vulnerability in the invscout command to execute arbitrary commands. IBM X-Force ID: 251207.

All versions of the package bwm-ng are vulnerable to Command Injection due to improper input sanitization in the 'check' function in the bwm-ng.js file. **Note:** To execute the code snippet and potentially exploit the vulnerability, the attacker needs to have the ability to run Node.js code within the target environment. This typically requires some level of access to the system or application hosting the Node.js environment.