Task Boundary Auditor
Security checks across malware telemetry and agentic risk
Overview
This skill is a markdown-only guidance skill that helps an agent recognize tasks beyond LLM capability and route them to humans or tools.
Install if you want the agent to be more conservative about verification, prediction, legal/medical/financial judgment, physical design, and similar high-stakes tasks. Expect it to add boundary-audit language and sometimes recommend human or specialized-tool review instead of answering directly.
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
VirusTotal
63/63 vendors flagged this skill as clean.