ClawHub
Back to skill

Security audit

体彩筛选

Security checks across malware telemetry and agentic risk

Overview

This skill is a lottery data analysis toolkit with disclosed, user-invoked CSV and optional public-data fetching behavior.

Before installing, be aware that the skill can read CSV files you point it at, write generated CSV/HTML/cache files, and optionally contact public lottery data sites via fetch_data.py. Treat lottery outputs as statistical references only, not betting advice.

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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill advertises and documents capabilities that imply file read/write and network access, but it declares no permissions or equivalent trust boundary. This creates a transparency and least-privilege problem: users and host platforms may invoke a skill that can access local files or remote sources without explicit disclosure, increasing the risk of unintended data access or exfiltration.

Description-Behavior Mismatch

Medium
Confidence
90% confidence
Finding
The document states that the tool does not include real-time fetching, yet elsewhere it explicitly includes online historical data retrieval. Contradictory statements about network behavior undermine informed consent and can cause operators to trust the skill in offline or restricted environments when it may actually initiate network access.

Intent-Code Divergence

Medium
Confidence
89% confidence
Finding
This is the same substantive issue as SDI-1 viewed through a direct contradiction lens: the skill claims no built-in fetching while documenting an online fetch feature. Such inconsistency is dangerous because it obscures network-enabled behavior, complicates policy enforcement, and weakens user ability to assess data-flow risk.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.