xiaobai-en

Security checks across malware telemetry and agentic risk

Overview

This instruction-only skill has no executable code, but it asks agents to archive youth/social-media discussion data and run public campaign activity without clear privacy or approval limits.

Install only if you intend to use it as a social campaign playbook and can enforce manual approval for posts, account access, and data collection. Avoid collecting identifiable data about minors or social-media users; prefer aggregated public metrics, define retention and deletion rules, and make output language follow the user's preference.

SkillSpector

By NVIDIA

Vulnerability Patterns

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
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Natural-Language Policy Violations

Medium

Confidence: 90% confidence
Finding: The skill explicitly prescribes bilingual Chinese/English output in its release rhythm and templates without indicating that the user's language preference should control responses. This can cause the agent to ignore user intent, reduce usability, and create policy/compliance issues in multilingual deployments where unsolicited language output may expose unnecessary content or confuse users.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal