ClawHub
Back to skill

Security audit

cn-blog-writing

Security checks across malware telemetry and agentic risk

Overview

This is a disclosed Chinese blog-writing assistant with no executable code, and its blog inspection and style-memory behavior fit its stated purpose.

Install this if you want writing help specifically for Chinese technical blog posts. On first use, provide only a public blog URL or sample posts you are comfortable having inspected, and be aware the agent may reuse the resulting style and SEO analysis in later writing tasks.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
85% confidence
Finding
The README states that after installation, common requests like writing, revising drafts, checking for 'AI tone', and choosing the next topic will automatically trigger this skill. That activation scope is broad enough to intercept ordinary writing-related prompts without a clear opt-in at run time, which can override user expectations and cause unintended behavior in unrelated contexts.

Natural-Language Policy Violations

Medium
Confidence
73% confidence
Finding
The skill is explicitly framed as a Chinese blog-writing tool and describes behavior that assumes Chinese output, without indicating that language selection is optional. This can cause unwanted language coercion when a user asks for writing help in another language, reducing user control and potentially affecting downstream accuracy or usability.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill explicitly instructs itself to visit and analyze the user's blog, including content layout, tone, and SEO structure, but it does not clearly warn the user that external content will be accessed and inspected. This can cause unintended browsing of user-controlled or third-party content and silent collection of stylistic or site-structure information without sufficiently informed consent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.