Security audit

跨境电商资讯日报

Security checks across malware telemetry and agentic risk

Overview

This skill is a transparent Chinese news-briefing helper for cross-border ecommerce and does not include hidden code, credential access, persistence, or destructive behavior.

Install this if you want Chinese summaries of current cross-border ecommerce news. Expect it to contact public external websites and verify important platform policy changes against the linked original sources.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

High

Confidence: 94% confidence
Finding: The trigger instructions are unusually broad and explicitly say to invoke the skill for generic phrases like '跨境电商' and to '不要 undertrigger,' which pressures over-invocation. This can cause the agent to route loosely related queries into web-fetching behavior unnecessarily, increasing exposure to irrelevant external content, privacy leakage through unneeded browsing, and degraded answer quality.

Natural-Language Policy Violations

Medium

Confidence: 80% confidence
Finding: The skill description says it will always produce a Chinese markdown brief without indicating that this should depend on the user's language or preference. While not directly enabling code execution or data exfiltration, this can override user intent, reduce transparency, and create unsafe or misleading behavior in multilingual contexts where the user may misinterpret summarized policy or compliance information.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.