Ai Agent Builder Zh
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only Chinese guide for building AI agents, with expected examples involving APIs, tool calling, and memory, but no executable or hidden behavior.
Safe to install as a guide. Before building from its examples, use least-privilege API keys, avoid storing unnecessary personal or business data, and require human approval before agents write to business systems, send messages, create tickets, or update shared documents.
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
64/64 vendors flagged this skill as clean.