ClawHub
Back to skill

Security audit

小红书中英双语爆款文案生成器

Security checks for vulnerabilities and agentic risk

Overview

This is a simple bilingual viral-content writing template with no code execution or persistence, though it may over-apply marketing-style formatting to broad writing requests.

Install this only if you want an assistant to format social content in a bilingual, engagement-focused marketing style. Be cautious using it for sensitive, regulated, or non-promotional topics because the default framing pushes hooks, emotional appeal, CTAs, and hashtags.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill hard-requires bilingual English and Chinese output ('response me both in english and chinese') without user choice, which can override user preferences and cause unnecessary disclosure, confusion, or formatting mismatch. While not directly malicious, forced dual-language output reduces user control and can create avoidable policy and usability issues when the user requested only one language or a concise response.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill activates on very broad triggers like 'create viral content for any topic' and 'promote products, services, or ideas,' which can capture many generic writing requests and route them into engagement-maximizing behavior without clear user intent. In context, this increases the chance of inappropriate invocation, persuasive optimization, or content shaping for sensitive topics where a narrower, explicit opt-in would be safer.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.