ClawHub
Back to skill

Security audit

chinese-ad-copywriter

Security checks across malware telemetry and agentic risk

Overview

This skill appears purpose-aligned, but its compliance check sends submitted ad-copy content to a disclosed external API.

Install only if you are comfortable sending ad copy, titles, keywords, and platform names to the listed Tencent Cloud Function endpoint. Avoid using it with confidential client drafts, unreleased campaigns, personal data, or regulated claims unless the publisher provides acceptable privacy, retention, and token-handling terms.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill advertises executable shell scripts and a live external API backend but does not declare permissions or clearly constrain when those capabilities may be used. In an agent environment, hidden or undeclared shell/network behavior can lead to unexpected command execution or data exfiltration, especially if the model decides to invoke `check.sh`/`generate.sh` automatically based on user input.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The trigger list contains broad phrases such as general ad-copy and compliance terms that could cause the skill to activate for loosely related requests outside its intended narrow scope. Over-broad activation increases the chance that an agent routes unrelated user content into this skill, potentially causing unnecessary external API calls, shell-script use, or language/platform-specific behavior the user did not request.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The script transmits the full ad copy, title, keywords, and platform metadata to a third-party remote API without any explicit notice, consent prompt, or warning at runtime. This creates a privacy and data-governance risk because users may paste unpublished marketing plans, client content, or regulated text into the tool without realizing it leaves the local environment.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.