ClawHub
Back to skill

Security audit

Antivirus

Security checks across malware telemetry and agentic risk

Overview

MoltGuard appears to be a real security integration, but it installs runtime tooling, sends security checks to an external Core service, and exposes/stores API credentials with too little user control and privacy detail.

Review before installing. This skill is not clearly malicious, but it gives a third-party guardrail service visibility into sensitive agent activity and creates local API credentials. Install only if you trust OpenGuardrails/MoltGuard, understand what data is sent to Core, avoid sharing `/og_claim` output in logged or collaborative chats, and confirm how to revoke credentials and remove the plugin.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill directs the agent to immediately install the plugin and then run a local file read as a demonstration, without clearly warning that these actions execute commands, access local filesystem content, and may trigger communication with an external service. In an agent context, this kind of imperative onboarding can normalize unsafe auto-execution and reduce user awareness and consent around potentially sensitive operations.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The onboarding flow states that API credentials are obtained and stored under a local path, but it does not warn that these are sensitive secrets or describe expected file permissions and handling requirements. This can lead to insecure secret storage practices, accidental disclosure, or overbroad access by other local users or processes.

Missing User Warnings

Low
Confidence
76% confidence
Finding
The uninstall instructions invoke a local script that removes configuration, plugin files, and credentials, but they do not clearly warn that this is a destructive action affecting local state. In an agent setting, omission of that warning can cause unintended deletion of security tooling and stored credentials if the command is run automatically or without informed user approval.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.