Skill Quality Evaluator

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:41 PM UTC
• Version:
  1.0.0

Overview

This is an instruction-only skill for evaluating other skills, with disclosed report-writing behavior that users should keep scoped to intended files.

Install is reasonable if you want a checklist-based skill evaluator. Before using batch mode, point it only at the skill directory you intend to review, and avoid saving reports that include secrets, proprietary skill text, or sensitive internal evaluations unless persistent memory storage is acceptable.

SkillSpector

By NVIDIA

Vulnerability Patterns

• 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
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (2)

Missing User Warnings

Medium

• Confidence: 90% confidence
• Finding: The skill instructs the agent to save evaluation reports into a persistent memory path without warning the user or requiring confirmation. This can cause unintended modification of stored project data, create data retention/privacy issues, and let routine evaluation actions leave durable side effects that the user may not expect.

Missing User Warnings

Medium

• Confidence: 92% confidence
• Finding: The batch workflow directs saving a summary report to a memory path, again without making the persistent write behavior explicit. In batch mode, the risk is amplified because repeated automated writes can accumulate sensitive metadata about many skills/projects and overwrite or clutter shared memory locations.

VirusTotal

66/66 vendors flagged this skill as clean.