Security audit
ZM 高质量产出流程
Security checks across malware telemetry and agentic risk
Overview
This skill is a workflow and checklist package for managing formal content production, with no executable code or hidden data access detected.
This appears safe to install as a process-guidance skill. Before using it, understand that it may guide your agent to create local delivery-package files, route tasks to other agents, and prepare publishing materials; keep the confirmation gates in place for any external upload, send, delete, or payment action.
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.