Caring

Security checks across malware telemetry and agentic risk

Overview

This is a text-only quality-review skill that may activate on broad review requests but does not run code, access private data, or change anything on its own.

Install this if you want direct, structured quality feedback. Be aware that generic review requests may invoke this particular review style, and its tone is intentionally blunt; invoke it explicitly when that is the feedback mode you want.

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: 95% confidence
Finding: The skill advertises very broad trigger phrases such as "review this," "check my work," and "quality review," which are common, high-frequency requests that could cause this skill to activate in many unrelated contexts. That increases the chance of unintended interception or override of more specialized skills, leading to misrouting, inconsistent behavior, or application of this review lens where it is not appropriate.

Vague Triggers

Medium

Confidence: 95% confidence
Finding: The skill declares very broad trigger phrases such as "review this," "check my work," and "quality review," which are common in ordinary conversation and likely to match many unrelated requests. This can cause unintended activation of the skill, leading the agent to apply the wrong workflow, override more appropriate task handling, or produce unexpected review-oriented outputs in contexts where the user did not explicitly request this skill.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal