Onboarding Cro

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only onboarding optimization skill with no code, installs, credentials, persistence, or account-modifying behavior.

Installers should understand that the skill may use local product marketing context to tailor recommendations. Review that context for accuracy, but the artifacts show no hidden execution, credential handling, or privileged behavior.

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 (1)

Vague Triggers

Medium

Confidence: 86% confidence
Finding: The eval explicitly rewards triggering on very casual language like 'users sign up but then never come back,' which broadens activation beyond narrowly scoped onboarding requests. In an agent-routing context, this can cause the skill to capture generic retention or re-engagement conversations that may belong to other skills, increasing misrouting risk and making the system easier to steer with ambiguous prompts.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal