Data Engineering

Security checks across malware telemetry and agentic risk

Overview

This skill appears to be a data-engineering reference and workflow guide, with no evidence of hidden execution, credential access, persistence, or data exfiltration.

Install this as an advisory data-engineering knowledge skill. Avoid pasting secrets, credentials, production connection strings, or confidential architecture details into prompts unless you intentionally want the agent to analyze them, and treat generated SQL/YAML/templates as examples to review before use.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (2)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The activation phrase "Help with [Spark/Airflow/dbt/Kafka] issue" is broad enough to match generic support requests and can cause the skill to activate outside a clearly bounded data-engineering context. Over-broad triggers increase the chance of unintended routing, where users seeking general troubleshooting may invoke this skill and receive inappropriate or over-privileged guidance.

Vague Triggers

Medium
Confidence
91% confidence
Finding
The phrase "Audit our data infrastructure" is ambiguous and could match a wide range of requests, from architecture review to security assessment to operational troubleshooting. This ambiguity can lead to accidental activation and unsafe task expansion, especially because "audit" may imply authority to evaluate sensitive systems, configurations, or compliance posture.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal