glm-output
Security checks across malware telemetry and agentic risk
Overview
This skill is a narrow local GLM output-processing guide with no hidden network, credential, persistence, or installation behavior.
Reasonable to install if you need help processing GLM output. Before running the example code, verify the local file paths, required Python packages, lake depth, and hard-coded simulation start date against your own model files; those are correctness considerations rather than security concerns.
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.