GLM Autoroute
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only model-routing skill with disclosed delegation and file-writing behavior, but users should be aware it may create local memory and report files.
Install this only if you want complex tasks automatically delegated to GLM-5 and are comfortable with local MEMORY.md/research/report files being created. Avoid including secrets in prompts that may be forwarded to a spawned session, and review generated code files before running them.
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.