ClawHub
Back to skill

Security audit

Sol Tool Spotlight

Security checks across malware telemetry and agentic risk

Overview

This skill openly automates publishing AI tool reviews, but it schedules GitHub pushes using a local API key without enough user control or permission clarity.

Install only if you intentionally want scheduled, automatic blog publication from this machine. Review the external source script first, keep the MiniMax key narrowly scoped, run it manually before enabling launchd, and require a separate review step before any git push.

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 (3)

Context-Inappropriate Capability

Medium
Confidence
89% confidence
Finding
The skill documentation states it will auto-commit and push content to GitHub, which is a side effect beyond simple content generation and is not reflected in the declared permissions. This creates a trust and capability mismatch that can lead to unauthorized repository modification or hidden publication behavior if users rely on the manifest to understand risk.

Scope Creep

High
Confidence
95% confidence
Finding
The documented setup requires access to a local secret file and implies GitHub interactions not captured by the listed permissions of http.request and file.write. This underdeclared access model is dangerous because it obscures secret handling and outbound publication behavior, increasing the chance of credential misuse, unintended data exposure, or execution in a more privileged environment than users expect.

Missing User Warnings

Medium
Confidence
90% confidence
Finding
The description presents the skill as a harmless weekly mini-review generator but omits an explicit warning that it writes files and auto-pushes commits to GitHub. This can mislead users about the operational impact of invoking the skill, making accidental content publication or repository changes more likely.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal