Skylv Agent Evaluator

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:40 PM UTC
• Version:
  1.0.2

Overview

This is a simple local tool for scoring agent outputs, with the main caution that users may provide sensitive conversation logs.

Install only if you want a lightweight local heuristic scorer for agent logs. Before evaluating conversation history, provide only the minimum needed excerpt and redact secrets, credentials, personal data, proprietary information, and prompt material you do not want included in the current agent context.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The quick-start trigger phrase is generic enough that it could activate in ordinary conversation without clear user intent to invoke this skill. Because the skill evaluates conversation history, accidental invocation could cause unintended processing of prior chat content or produce unsolicited judgments about an agent's behavior.

VirusTotal

65/65 vendors flagged this skill as clean.