Claude Code Architecture
Security checks across malware telemetry and agentic risk
Overview
This skill is a disclosed reference pack for AI-agent architecture patterns, with no hidden install behavior or automatic high-risk actions found.
Reasonable to install as reference material. Before allowing an agent to write generated templates into a real project, confirm the destination path, review the diff, and test the changes in a branch or sandbox.
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
63/63 vendors flagged this skill as clean.