Variant Pathogenicity Predictor

Security checks across malware telemetry and agentic risk

Overview

This is a simple local Python scoring helper for variant pathogenicity, with no evidence of network access, credential use, persistence, or hidden behavior.

Install only if you are comfortable running a local Python script. Treat its output as a basic heuristic, not medical or diagnostic advice, and avoid placing identifiable genomic or clinical data in shared, synced, or retained workspaces unless you have appropriate access controls.

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

Findings (1)

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The skill explicitly supports reading variant inputs and writing reports to workspace files, but it does not clearly warn that genomic variant data and gene associations may be sensitive and potentially identifiable. In environments where workspace contents are retained, shared, logged, or synced, this can lead to unintended exposure of sensitive biomedical data even without any overtly malicious behavior.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal