Security audit

NIBVOK AI Security

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed OpenClaw security plugin that inspects tool calls, prompts or blocks risky actions, and records an audit log.

Install this only if you want an enforcement layer that can block or require approval for agent actions, including some attempts to disable or reconfigure the plugin. Review the audit-log location and policy presets first, because command metadata may be recorded and incorrect policy settings can interrupt normal work.

SkillSpector

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

Static analysis

Detected: suspicious.dangerous_exec, suspicious.env_credential_access

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: test-audit-log-isolation.mjs:50
Evidence: execFileSync(process.execPath, [join(HERE, suite)], { cwd: HERE, env, stdio: "pipe" });

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: test-classifier.mjs:454
Evidence: { command: "node -e \"require('child_process').execSync('openclaw config set x y')\"" }, "deny");

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: test-policies.mjs:105
Evidence: const raw = execFileSync(

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: classifier.js:102
Evidence: export const ARTIFACT_ROOT = process.env.ASF_ARTIFACT_ROOT || "/var/asf/artifacts/";