ClawHub
Back to skill

Security audit

AI人味写作技

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only writing skill with no executable code, but it is explicitly designed to make AI-generated text harder to detect as AI-written.

Install only if you intentionally want a Chinese creative-writing rewrite assistant and understand that it is built around making text look less AI-generated. Do not use it to misrepresent authorship or evade school, workplace, or platform disclosure rules. Apply it only to text you choose and review any edits before publishing or saving over originals.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list includes broad terms such as '中文写作', '润色', and '故事改写', which are common in normal writing requests and can cause the skill to activate when the user did not specifically ask for this workflow. Over-broad activation is risky because the skill has write/edit capabilities and is designed to substantially transform text, increasing the chance of unintended modification or unexpected behavior.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill declares Read, Write, and Edit permissions but does not clearly warn users that it may rewrite or alter their text. In context, the whole skill is a multi-stage rewriting engine, so silent modification is especially concerning because users may expect analysis or suggestions, not direct content changes.

Natural-Language Policy Violations

Medium
Confidence
83% confidence
Finding
The skill is scoped entirely around Chinese-language creative writing and does not offer a user choice or justify language restriction at the point of instruction. This is less severe than the other findings, but it can still create unsafe or misleading behavior if the skill activates on non-Chinese requests and transforms content into an unintended language or style.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.