Java Maven Code Review

Security checks across malware telemetry and agentic risk

Overview

This is a local Java/Maven review helper that reads a target project and writes a report, with no evidence of hidden data transfer, destructive behavior, or credential misuse.

Install only if you are comfortable letting the agent read the Java/Maven project you provide and create report files in your workspace. Treat the bundled script’s output as a first-pass heuristic review, not a complete assurance that naming, duplication, module-boundary, or maintainability issues were fully checked.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
93% confidence
Finding
The skill promises review of ZIP archives and GitLab repositories and claims substantive Java/Maven analysis, but the implementation behavior described by the finding appears to rely on precomputed local inputs and superficial keyword checks. This mismatch can mislead users into trusting a report that did not actually inspect the supplied source, creating integrity risk, false assurance, and potential omission of serious code or configuration problems.

VirusTotal

67/67 vendors flagged this skill as clean.

View on VirusTotal