Clarity Gate

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent document-quality checker for RAG preparation, with local hashing helpers and no evidence of hidden or harmful behavior.

Install is reasonable for RAG pre-ingestion document review. Use explicit invocations when possible, review generated CGD/HITL outputs before adding them to a knowledge base, and run the helper scripts only on files you intentionally want analyzed.

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: 72% confidence
Finding: Several trigger phrases are broad, such as requests about whether an LLM can read something safely or generic review-for-clarity prompts. Broad triggers increase the chance of accidental activation on unrelated document-review tasks, which may steer workflows into this skill's specialized format, HITL, and ingestion logic unexpectedly.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal