AI 写作诊断
Security checks across malware telemetry and agentic risk
Overview
No malicious behavior is evident from the supplied scan signals or the available workspace context.
This can be treated as benign based on the available evidence, but review the actual skill files at install time if available, especially any declared credentials, network access, or commands it asks the agent to run.
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
52/52 vendors flagged this skill as clean.