Security audit
ai视频人物真实性方法论
Security checks across malware telemetry and agentic risk
Overview
This is a documentation-only AI video prompt guide with no code execution or system access, though its realism tips should be used responsibly.
Installing this skill appears technically low risk because it is prompt-writing reference material only. Users should be mindful that it includes techniques for making AI-generated people look realistic, so synthetic media should not be used for impersonation or presented as real without appropriate disclosure.
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
65/65 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.