ClawHub
Back to skill

Security audit

Mc Mod Translate, English to Simplified Chinese

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Minecraft mod translation helper that uses a local dictionary plus clearly described online lookups, with no hidden persistence or credential access found.

Install only if you are comfortable with downloading a large community-maintained SQLite dictionary from GitHub and sending individual lookup terms to zh.minecraft.wiki during fallback queries. Use --no-wiki for local-only lookups, and consider pinning or verifying the dictionary source if reproducibility matters.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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
Findings (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill instructs the agent to perform network operations such as downloading a SQLite database from GitHub and querying zh.minecraft.wiki, but no permissions declaration is present. This creates an undeclared trust boundary: users or orchestrators may assume the skill is offline/local while it can exfiltrate prompts, retrieve untrusted remote content, or fail unpredictably in restricted environments.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The documented behavior materially overstates what the skill does: it claims end-to-end translation capability and context-based fallback, but the described implementation mainly downloads/queryies a dictionary and wiki data. This mismatch is dangerous because operators may trust it to safely transform files or keep processing local, while the actual behavior includes remote fetching and lacks the promised translation logic, increasing the chance of unsafe automation decisions and data exposure.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.