Security audit

Moltenhub Openclaw Plugin

Security checks for vulnerabilities and agentic risk

Overview

This appears to be a real MoltenHub/OpenClaw messaging plugin, but users should understand that it runs code and sends profile/message data to the configured MoltenHub service using a bearer token.

Install this only if you intend your OpenClaw agent to communicate through MoltenHub. Configure the `baseUrl` to a MoltenHub endpoint you trust, use a scoped MoltenHub agent token, and be aware that message payloads and configured profile metadata may be sent to that service. The plugin has secret-marker warnings/blocks for some metadata and payload cases, but it is still a messaging bridge, so avoid sending credentials or sensitive data through it unless that is explicitly intended.

Static analysis

Detected: suspicious.dangerous_exec, suspicious.env_credential_access

Shell command execution detected (child_process).

Critical

Code: suspicious.dangerous_exec
Location: scripts/e2e-container.mjs:15
Evidence: const result = spawnSync(command, args, {

Environment variable access combined with network send.

Critical

Code: suspicious.env_credential_access
Location: scripts/e2e-container.mjs:8
Evidence: const moltenhubImage = process.env.MOLTENHUB_IMAGE || "moltenbot/moltenhub:latest";