Security audit

regulation-monitor / 监管动态追踪

Security checks across malware telemetry and agentic risk

Overview

The skill is a disclosed crawler for public Chinese regulatory updates, with no hidden persistence, credential access, or unrelated data handling found.

Install this only if you are comfortable with the agent making live requests to the listed public regulator websites. Use explicit regulator and day-range options when possible, and only configure a proxy you trust.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
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 invoke a network-capable crawler and even supports custom proxy configuration, but the manifest declares no corresponding permissions. This creates a transparency and governance gap: users or platforms may not realize the skill can make outbound requests to multiple external regulatory sites, which can bypass expected review or consent controls.

Vague Triggers

Medium

Confidence: 82% confidence
Finding: The trigger phrases are broad and overlap with ordinary requests about recent news, policies, or enforcement, increasing the chance the skill is auto-invoked when the user did not intend to run a crawler. Because the skill performs live network retrieval, accidental invocation can cause unexpected outbound traffic, privacy issues around user prompts, or retrieval from untrusted remote content.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.