Security audit

omni-expert

Security checks across malware telemetry and agentic risk

Overview

This is a broad general-purpose consulting skill with disclosed tool-use guidance, but no executable code, hidden install behavior, persistence, or evidence of data exfiltration.

Install this only if you want a broad catch-all assistant workflow. Review or override its actions when tasks involve private files, credentials, code execution, network searches, financial/legal decisions, or changes to external systems.

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: 95% confidence
Finding: The skill’s trigger conditions are extremely broad (e.g. '万能', '全能', '帮我搞定') and overlap with common user phrasing, which can cause unintended invocation for many ordinary requests. That increases the chance that a highly permissive, tool-using generalist skill takes control when a narrower or safer skill should apply, expanding access to file, code, or network actions without clear user intent.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The workflow explicitly encourages use of file operations, code execution, and network search, but the skill does not warn users about possible system changes, data access, privacy exposure, or external communications. In a universal skill that positions itself as 'do anything' and 'actively take charge,' this omission can normalize risky actions without informed consent or clear boundaries.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.