Back to skill

Security audit

技能合并器

Security checks across malware telemetry and agentic risk

Overview

This skill is a plain-language workflow for combining other skills and does not contain executable code, hidden persistence, credential handling, or network behavior.

Install this if you want help consolidating overlapping skills, but review any merged skill it produces before using or publishing it because the output can change how future agent workflows are triggered and routed.

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
93% confidence
Finding
The trigger condition '合并关键词' is underspecified, so the skill may activate on broad or ambiguous user inputs that merely mention merging or combination. In a skill that rewrites or consolidates other skills, unintended invocation can cause incorrect routing, accidental modification guidance, or interference with neighboring skills, making this a real prompt-scope vulnerability rather than a harmless wording issue.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.