释空悟之家
Security checks across malware telemetry and agentic risk
Overview
This instruction-only skill is transparent and has no code, but it sets a standing multi-agent coordination role that can spawn current and future subagents without clear user approval or data-sharing limits.
Review before installing if you do not want a skill to influence every conversation or coordinate subagents automatically. Use it only with an explicit expectation that subagents may be spawned, and require clear approval before sharing sensitive task context with any other agent.
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
51/51 vendors flagged this skill as clean.