Data Quality Validator

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only data quality auditing skill whose data access is expected for its purpose, but users should scope it carefully around sensitive datasets.

Install with normal caution for data tools. Point it only at specific approved projects or datasets, prefer masked or sampled data when possible, and review generated reports before sharing because profiles can include sensitive values, internal schemas, and business metrics.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
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: 88% confidence
Finding: The skill explicitly instructs the agent to read data source definitions, pipeline code, schema files, and sample data, which in real environments can include production records, credentials, internal schemas, and personal data. Because the skill provides no data-minimization, redaction, or least-privilege guidance, it creates a meaningful risk of unnecessary exposure of sensitive information during routine use.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal