Ollama Model Tuner
Security checks across malware telemetry and agentic risk
Overview
This skill appears to run a small local Ollama prompt-evaluation script, with some overstated tuning claims but no evidence of hidden data theft, persistence, or destructive behavior.
Before installing, confirm the source of the runnable command and the Python ollama package. Treat this as a lightweight local prompt-testing helper, not a full LoRA or model fine-tuning tool, and only point it at datasets you are comfortable sending to your Ollama service.
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.