Security audit

Java健壮编程

Security checks across malware telemetry and agentic risk

Overview

The skill is mainly Java coding guidance, but it asks for broad read/write/command authority and describes callbacks, API use, and automation beyond that stated purpose.

Review this skill before installing. Its Java guidance is ordinary, but its declared authority and documentation are broader than a Java robustness helper needs. Only install it if you are comfortable granting read/write/exec capability and possible API or callback use, and require explicit approval before it modifies files, runs commands, or sends data outside the workspace.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 93% confidence
Finding: The skill is advertised as narrowly focused Java robustness guidance, but the later documentation expands it into generic file processing, API integration, and command execution. This capability mismatch increases the chance the agent will invoke a higher-privilege skill for requests that appear low-risk, enabling unnecessary write/exec/network actions beyond user expectations.

Context-Inappropriate Capability

Medium

Confidence: 90% confidence
Finding: A callback URL and external API/network-oriented behavior are not justified by a Java robustness coding assistant and introduce unnecessary outbound communication paths. In an agent context, this can enable data exfiltration, SSRF-style access to internal endpoints, or silent transmission of code and metadata to third parties.

Intent-Code Divergence

Medium

Confidence: 92% confidence
Finding: The documentation frames the skill as Java robustness help but later describes a broad automation tool capable of files, APIs, commands, and data pipelines. This hidden breadth is dangerous because users and orchestrators may grant trust based on the narrow label while the skill can perform materially more sensitive operations.

Vague Triggers

Medium

Confidence: 84% confidence
Finding: The activation wording is broad enough to match generic coding, debugging, deployment, and development requests, which can cause over-selection of a skill that has read/write/exec capabilities. Over-broad routing increases the likelihood of this skill being invoked in contexts where those powerful tools are unnecessary, raising the attack surface.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The markdown advertises file writing, external API use, and command execution without prominent warnings or consent requirements for these impactful actions. In an agent system, that omission can normalize dangerous behavior and permit destructive filesystem changes, data leakage over the network, or shell command abuse under the guise of a benign programming helper.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.