OpenLang
Security checks across malware telemetry and agentic risk
Overview
This is a text-only shorthand protocol for agent-to-agent messages, with no code, installer, credentials, persistence, or automatic execution behavior.
Install only if you want compressed sub-agent communication. Keep normal user approval and policy checks in place: OpenLang messages should not be treated as authorization to run commands, modify files, access databases, use credentials, or perform destructive actions unless separately allowed.
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.