Security audit
代码评审
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only code review skill that asks the agent to inspect user-provided project documents and code, with no hidden execution, persistence, credentials, or external data transfer shown.
Install only if you want a structured Chinese-language code review workflow. When using it, share only the repository files, requirements, clarifications, and design documents that are relevant to the review, because the skill explicitly asks the agent to read those materials.
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.
