ClawHub
Back to skill

Security audit

Grok Build

Security checks across malware telemetry and agentic risk

Overview

This skill is a short, disclosed guide for installing and using xAI's Grok Build coding CLI, with no hidden files or automatic behavior found.

Before installing, review the remote installer commands and run Grok Build with the least privilege practical for your project. Treat it like any coding agent: it can edit files, run shell commands, use the network, and may use authenticated sessions after browser login.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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
Findings (1)

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill explicitly promotes shell execution, file editing, web search, and headless/CI operation, but it does not warn users that these capabilities can modify local files, execute destructive commands, access sensitive data, or trigger unintended network activity. In an agent-skill context, omission of these safety caveats increases the chance that users will run the tool with excessive trust or automation, especially in unattended environments.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.