ClawHub
Back to skill

Security audit

Java手册

Security checks for vulnerabilities and agentic risk

Overview

This Java manual skill is mostly documentation, but it asks for broad write and command-execution authority without clear limits.

Review this skill before installing. It does not contain obvious malicious instructions, but its requested write and command-execution permissions are broader than a Java manual normally needs. Use it only if you are comfortable granting those tools and can supervise any file changes or commands it suggests.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Confidence
93% confidence
Finding
The skill is presented as a passive Java manual, but later advertises operational capabilities such as file writing, API integration, and command execution. That mismatch can mislead users and agents into authorizing side-effecting behavior they would not expect from a reference skill, increasing the chance of unsafe execution or overbroad trust.

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
The top-level description frames the artifact as a Java development manual/reference, while also implying active code-generation, debugging, deployment, and workflow use. This contradiction weakens informed consent and can cause downstream systems to classify the skill as low-risk reference content when it may trigger higher-risk actions.

Vague Triggers

Medium
Confidence
87% confidence
Finding
The invocation guidance is overly broad, covering generic development tasks like code generation, debugging, testing, and deployment. Broad triggers can cause the skill to activate in many common contexts, potentially introducing unnecessary tool use or side effects without the user specifically requesting this skill.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill advertises file writes, external API calls, and command execution, but does not pair those capabilities with prominent warnings about side effects, data exposure, or consent requirements. In an agent environment, that omission can lead users to invoke a seemingly harmless handbook skill that performs state-changing or networked actions unexpectedly.

Static analysis

No suspicious patterns detected.