Security audit
导出治理与审批 (Agentic AI 科研平台)
Security checks across malware telemetry and agentic risk
Overview
This skill is a local export-governance helper whose data access and localhost status reports match its stated purpose.
Install this only for a local research/export-governance workflow where `/home/ubuntu/workspace/demo/mock_data/governance.json` and `localhost:5001/api/report` are expected. Review whether approval and audit details should be shown to the agent or user in your environment.
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.