Ai Content Polish

Security checks across static analysis, malware telemetry, and agentic risk

Overview

This skill is markdown-only and disclosed, but its core purpose is to make AI-written Chinese text harder to recognize as AI-generated.

Install only if you want a Chinese writing-style reviewer and are comfortable with its AI-detection-evasion framing. Avoid using it to bypass academic, platform, publisher, or workplace disclosure rules; a safer version would focus on clarity, tone, and readability without hiding AI authorship.

SkillSpector (3)

By NVIDIA

Vague Triggers

Medium

Confidence: 90% confidence
Finding: The activation cues are broad enough to match ordinary editing and polishing requests, which can cause the skill to be invoked in contexts the user did not explicitly intend. Because the skill is designed around removing AI-writing signals and '反检测' use cases, overbroad triggering increases the chance of steering benign writing assistance into evasion-oriented behavior.

Ssd 2

Medium

Confidence: 98% confidence
Finding: The description explicitly includes '反检测', 'AI watermark', and 'AI content detection' as triggers, framing the skill as a tool to make AI-generated text harder to identify. That is dangerous because it facilitates evasion of platform integrity controls, academic or publishing disclosure rules, and downstream moderation systems designed to identify synthetic content.

Ssd 2

Medium

Confidence: 99% confidence
Finding: The tagline '让 AI 写的东西看起来不像 AI 写的' directly states an evasion objective in plain language. In context, this is not incidental wording but the core purpose of the skill, making the skill materially more dangerous because its primary function is to conceal the origin or detectability of AI-generated content.

Static analysis

No static analysis findings were reported for this release.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal