射频领域客户项目技术支持

Security checks across malware telemetry and agentic risk

Overview

This is a Chinese RF customer-support playbook with no code or install-time behavior, though users should handle customer/project details as confidential.

Install is reasonable for RF customer-support workflows. Avoid putting real customer names, roadmaps, complaint histories, test data, or design documents into general-purpose agent memory or shared notes unless your organization permits it.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (1)

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The invocation description is broad enough to overlap with general customer-support and RF advisory requests, which can cause the skill to trigger outside its intended scope. In an agent system, overbroad routing can expose internal workflows, produce irrelevant or overly authoritative answers, and create prompt-surface expansion that increases the chance of downstream misuse or policy bypass.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal