Security audit
Taste Kafka
Security checks across malware telemetry and agentic risk
Overview
This is a literary style-reference skill with no code, credentials, persistence, or sensitive system access.
Installers should expect this skill to influence the agent's creative writing and design style toward Kafka-like themes. The only notable consideration is that it contains external InspiredHub demo/contact links, but the skill does not instruct automatic access to them or request sensitive permissions.
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.
Static analysis
No suspicious patterns detected.