ClawHub
Back to skill

Security audit

llm-verify

Security checks across malware telemetry and agentic risk

Overview

The skill’s core purpose is coherent, but its install instructions rely on remote code execution and a default installer that can modify every detected AI coding tool.

Install through ClawHub or a scoped/project-local target where possible. Avoid running the remote shell or PowerShell installer unless you first inspect the fetched script and verify the source you intend to trust. If using llm-verify, provide only the endpoint and API key you intend to test, and prefer a temporary or limited API key because the tool will exercise that credential against the selected endpoint.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
95% confidence
Finding
The README instructs users to execute a remote shell script directly via curl-pipe-to-sh, which bypasses inspection and integrity verification. If the upstream repository, network path, or hosting account is compromised, users could immediately run attacker-controlled code on their machine.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The Windows install command uses Invoke-RestMethod piped into iex, which executes fetched PowerShell content immediately without review or authenticity checks. This creates the same remote-code-execution risk as the Unix variant, with potential for full user-context compromise if the script source is malicious or tampered with.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The documentation says the built-in installer writes to every detected AI coding tool, which can modify multiple environments and trust boundaries in one action. In a skill context, broad auto-installation increases the blast radius of mistakes or compromise because it propagates configuration or executable content into several developer tools at once.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.