Model Guard

Security checks across malware telemetry and agentic risk

Overview

This skill openly does what it claims: it checks OpenClaw model quotas and automatically changes the default model according to fixed rules.

Install or schedule this only if you want OpenClaw's default model to change automatically. Review guard.js first to confirm the candidate model list, 20% threshold, fallback model, and the assumption that some Gemini Anti-Gravity models count as full quota when not reported.

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 (1)

Missing User Warnings

Medium

Confidence: 96% confidence
Finding: The script automatically runs `openclaw models set ...` to change the user's configured model based on parsed quota output, without any prior confirmation, explicit consent flow, or dry-run mode. In an agent skill context, silently mutating user configuration can alter behavior, cost, privacy characteristics, and downstream trust assumptions, especially because the change is triggered by external command output rather than direct user intent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal