Xungen

Security checks across malware telemetry and agentic risk

Overview

This is a prompt-only entertainment/cultural-history skill that uses surnames or hometowns to generate playful Shang-dynasty ancestry-style output, with no code execution or external data access.

Install only if you are comfortable with the agent using a surname or hometown to produce playful ancestry-style interpretations. Treat outputs as entertainment and cultural commentary rather than factual genealogy, and prefer explicit '/xungen' use if you want to avoid accidental activation.

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 (2)

Vague Triggers

Medium

Confidence: 93% confidence
Finding: The skill advertises activation on very broad phrases like discussions of surnames, ancestry, family history, and 'I am from X,' which are common in ordinary conversation. This can cause unintended invocation, leading the agent to switch tasks unexpectedly and potentially elicit or process sensitive personal background information without clear user intent to use this specific skill.

Vague Triggers

Medium

Confidence: 89% confidence
Finding: The usage guidance says the skill can activate when a user says broad phrases such as '我姓X' or '帮我查查祖先', which are not sufficiently specific to intentional tool use. In multi-purpose chats, this increases the chance of accidental activation and unsolicited inference from personal identifiers like surname and hometown.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal