ClawHub
Back to skill

Security audit

Java Toolkit Free

Security checks across malware telemetry and agentic risk

Overview

This is mostly a Java checklist skill, but it asks for broader write/exec-style authority and describes create, modify, delete, export, and save operations that do not fit its stated purpose.

Review this skill before installing. It appears to be a Java coding checklist rather than malware, but its requested write/exec capability and vague data-affecting operation language should be narrowed or clarified before routine use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Intent-Code Divergence

Medium
Confidence
93% confidence
Finding
The skill claims generic create/query/modify/delete/export/save capabilities driven by parameters, but the rest of the document presents a passive Java checklist tool. This mismatch can cause an agent to over-grant the skill authority, invoke write or exec behaviors outside the user’s intent, or treat arbitrary inputs as operational commands rather than analysis content.

Vague Triggers

Medium
Confidence
90% confidence
Finding
The trigger condition covers broad categories like code generation, programming assistance, debugging, testing, and deployment, which are common requests far beyond a Java checklist tool’s stated scope. Overbroad routing criteria can cause the agent to activate this skill in unrelated situations, especially because the skill declares read/exec/write tools, increasing the chance of unnecessary command execution or file modification.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
Advertising modify/delete/save/export operations without clear warnings, confirmation requirements, or scope restrictions is risky, especially in a skill that has write and exec tools available. An agent may infer permission to alter files or export data as part of normal operation, turning an advisory checklist into a data-affecting automation path.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.