Ro Prompt Optimizer

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

Overview

This is a coherent prompt-optimization skill with optional local save and clipboard features, and the artifacts do not show hidden access, exfiltration, destructive behavior, or unsafe automatic execution.

Install this only if you want a Chinese-language R-O prompt rewriting helper. Avoid entering secrets, credentials, private business data, or sensitive personal information if you plan to print, save, or copy the result, and use the optimized prompts only for legitimate safe tasks.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• 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

Findings (4)

Description-Behavior Mismatch

Medium

• Confidence: 80% confidence
• Finding: The skill claims to optimize prompts, but it also writes user-derived content to local files. Even though saving is user-selected, this creates an unnecessary data persistence channel that can store sensitive prompts or generated content on disk without clear disclosure, increasing privacy and data-handling risk.

Context-Inappropriate Capability

Low

• Confidence: 74% confidence
• Finding: Clipboard access copies optimized prompts into the system clipboard, which is a shared OS resource and may expose sensitive prompt contents to other applications, clipboard history managers, or later accidental pastes. This is not overtly malicious, but it is an unnecessary side effect for a prompt optimizer unless clearly disclosed and justified.

Ssd 4

Medium

• Confidence: 90% confidence
• Finding: The optimizer systematically wraps arbitrary user prompts in stronger role, context, audience, and output framing without any safety screening. In a prompt-optimization skill, this materially increases the effectiveness and persuasiveness of harmful or policy-evasive prompts, making unsafe requests easier to operationalize.

Ssd 4

Medium

• Confidence: 88% confidence
• Finding: The usage guidance explicitly tells users to apply optimized prompts directly in AI interactions and iteratively refine them. In context, this encourages operational use of any optimized prompt, including malicious ones, and makes the skill more dangerous because its stated purpose is to improve prompt efficacy rather than enforce safety boundaries.

VirusTotal

67/67 vendors flagged this skill as clean.