Security audit

监控器

Security checks for vulnerabilities and agentic risk

Overview

This looks like a generic ping-monitor skill, but it asks for broad command execution and activation scope beyond simple ICMP checks.

Install only if you are comfortable with a skill that can use shell execution and whose instructions are broader than simple ping checks. Prefer a version that limits execution to explicit network reachability commands, validates targets, and narrows activation guidance to ICMP monitoring.

SkillSpector

Vulnerability Patterns

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
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (4)

Intent-Code Divergence

Medium

Confidence: 94% confidence
Finding: The skill is documented as a narrow ICMP health-check monitor, but it declares the exec tool and a Markdown+execute classification, which materially expands it into arbitrary command execution. This mismatch can cause an agent or user to invoke a much more privileged capability than expected, increasing the chance of command abuse, lateral actions, or execution of unsafe shell operations under a benign-looking monitor label.

Context-Inappropriate Capability

Medium

Confidence: 96% confidence
Finding: For an ICMP health-check skill, arbitrary execution is not justified by the described functionality and creates unnecessary privilege. A caller may be able to turn a simple monitoring workflow into shell execution, enabling system reconnaissance, file access, or execution of additional programs beyond the intended ping checks.

Vague Triggers

Medium

Confidence: 86% confidence
Finding: The usage guidance is overly broad, including generic system monitoring, log analysis, alerting, deployment management, and workflow automation language that goes well beyond ICMP checks. This can cause the skill to be selected in unrelated contexts, exposing its exec capability more often and increasing the attack surface through accidental or inappropriate invocation.

Vague Triggers

Medium

Confidence: 83% confidence
Finding: The additional guidance is vague and generic enough to overlap with common efficiency and automation requests, which does not align with a narrowly scoped monitoring skill. In context, that ambiguity is more dangerous because the skill also has exec available, making misrouting to a privileged skill more likely.

Static analysis

No suspicious patterns detected.