Security audit
Test Time Compute Guide
Security checks across malware telemetry and agentic risk
Overview
This is a documentation-only guide for LLM reasoning techniques and does not request sensitive access or run code automatically.
Reasonable to install as a guide. If you implement the examples using third-party model providers, Transformers packages, or custom reward models, review those separate dependencies and avoid sending sensitive prompts or data to untrusted services.
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
66/66 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.