ClawHub
Back to skill

Security audit

产业-高成长赛道挖掘

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Chinese-language industry research report generator that uses web research and scoring tools without local execution, persistence, or credential access.

Install this if you want a Chinese-language, web-researched framework for industry-chain growth-track reports. Expect it to perform extensive web searches and produce a long, structured report; users who need another language or a lighter analysis should state that preference explicitly.

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
  • 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 (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill content strongly constrains behavior to Chinese-language output without any user opt-in, which can override the user's preferred language or the surrounding system's locale expectations. In this context, the risk is elevated because the skill is long, mandatory, and output-format rigid, making it likely to dominate the assistant's response style and reduce user control, accessibility, and correct routing in multilingual environments.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The trigger list is very broad and overlaps with many generic industry-research or planning requests, so the skill could activate outside the author's intended scope and override more appropriate skills or user preferences. In this skill's context, over-activation is more dangerous because the instruction block is highly prescriptive, forces a specific workflow, and compels real-time web use and detailed report generation, which can steer conversations unnecessarily and amplify downstream errors or unwanted behavior.

Content

No source excerpt is available for this finding.