ClawHub
Back to skill

Security audit

AI写作提示词与公众号爆款文案指南

Security checks across malware telemetry and agentic risk

Overview

This is an instruction-only Chinese writing guide with no code execution or credential access, but users should avoid using its copying prompts to reproduce others' content without permission.

Install only if you want a Chinese self-media writing workflow. Do not use the extraction prompts to copy or republish third-party articles, captions, subtitles, or posts unless you have rights; prefer summaries, analysis, original drafting, licensed material, and human review before publishing.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (2)

Context-Inappropriate Capability

Medium
Confidence
97% confidence
Finding
The skill explicitly instructs the agent to '100%还原' and fully extract complete third-party content across platforms, preserving titles, body text, emojis, tags, subtitles, and formatting. That goes beyond ordinary writing assistance and enables large-scale reproduction of copyrighted material or unauthorized scraping-like reuse, creating legal, policy, and trust risks; the surrounding workflow then encourages rewriting that copied material into publishable output.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The activation description is broad enough to match many generic writing-help requests, increasing the chance the skill is invoked outside its narrow intended context. When combined with later embedded prompts for copying and aggressive rewriting, overbroad triggering makes unsafe behavior more likely to surface unexpectedly in ordinary user sessions.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.