Data Cleaning & Annotation Workflow
Security checks across malware telemetry and agentic risk
Overview
This skill is a transparent dataset cleaning and upload workflow, with expected local file handling and external uploads that users should use deliberately.
Install only if you are comfortable using Kaggle and data.smlcrm.com for this workflow. Use public or approved datasets, avoid uploading sensitive data unless the platform is authorized for it, install Kaggle dependencies in a trusted environment, and run the downloader in a new empty folder so its ZIP extraction and cleanup do not affect unrelated files.
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.