Back to skill

Security audit

2026世界杯实时比分系统

Security checks across malware telemetry and agentic risk

Overview

This skill appears to deploy the advertised World Cup scores site, but its setup script pulls changing external code, runs it in the background, and can kill any service using port 3001.

Review the external GitHub repository before running this skill, preferably pin it to a known commit and run it in a container or disposable host. Avoid the one-click script on systems where port 3001 may be used by another service, and confirm how to stop the background Node process before installing. VirusTotal and the static scan were clean, so the concern is operational scope and supply-chain control, not confirmed malware.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (7)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill instructs users to run shell commands such as git clone, npm install, mkdir, and node/nohup, but no permissions or explicit warning framework is declared for that capability. This creates a trust and consent gap: a user may trigger deployment behavior that executes system-level actions and pulls unreviewed code from a third-party repository.

Description-Behavior Mismatch

Medium
Confidence
96% confidence
Finding
The script does not deploy the packaged skill contents; it clones and updates a hard-coded external GitHub repository at runtime. This creates a supply-chain and integrity risk because the executed code can differ from the reviewed skill, allowing unreviewed upstream changes or repo compromise to affect the host during installation.

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The script checks whether port 3001 is in use and then forcibly kills the process bound to that port. This can terminate unrelated services, causing denial of service or data loss, and is not safely scoped to only processes previously started by this application.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The activation phrase is very broad: asking to 'help deploy the 2026 World Cup live score system' is a natural-language request that could appear in ordinary conversation and unintentionally trigger deployment guidance. Because the skill performs infrastructure/setup steps, accidental activation could lead to network access, package installation, or server changes without sufficiently deliberate user intent.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill prominently relies on multiple third-party data sources and an external GitHub repository, but it does not warn users up front that deployment and runtime will access external networks. This lack of disclosure reduces informed consent and can expose users to privacy, compliance, or supply-chain risk in environments where outbound connections are restricted or monitored.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
Using fuser -k on the target port without an interactive prompt or strong warning makes the script destructive by default. A user running a 'one-click deploy' script may unexpectedly disrupt production or local services that happen to use the same port.

Session Persistence

Medium
Category
Rogue Agent
Content
node server.js

# 生产模式(后台守护)
nohup node server.js > server.log 2>&1 &
```

访问 `http://你的服务器IP:3001` 即可使用。
Confidence
84% confidence
Finding
nohup

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.