Security audit

Codex

Security checks for vulnerabilities and agentic risk

Overview

This is a broad Codex runtime integration plugin, but its sensitive session, auth, and control powers are disclosed, scoped, and gated by owner/operator controls.

Install this only if you want OpenClaw to integrate deeply with Codex. Review settings such as user-home access, raw transcript tools, write controls, native plugin authorization, approvalPolicy, sandbox mode, and danger-full-access before enabling them, because those choices affect access to local Codex sessions, credentials, and execution authority.

SkillSpector

SkillSpector was not run because this plugin release contains no bundled skills.

Static analysis

Detected: suspicious.dangerous_exec

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: dist/dynamic-tools-CmN5RAip.js:2041
Evidence: child = spawn(command, args, {

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: dist/transport-stdio-BGjHoYLr.js:103
Evidence: const inspector = execFile("ps", args, {