ClawHub
Back to skill

Security audit

AI KEY·这生意靠谁

Security checks for vulnerabilities and agentic risk

Overview

This is a business dependency checkup skill with no executable code or hidden privileged behavior, though it uses memory for scoped follow-up calibration.

Before installing, be comfortable sharing business dependency details such as suppliers, platforms, leases, licences, key staff, and payment channels. Review the memory behavior if those details are sensitive, since the skill records scoped lessons and switching-cost facts for later calibration.

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)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The trigger list includes very broad natural-language phrases such as general business-risk questions, which can cause the skill to activate during ordinary conversation without clear user intent. That increases the chance of unintended instruction takeover, context switching, or collection of business-sensitive details under the wrong workflow.

Natural-Language Policy Violations

Medium
Confidence
93% confidence
Finding
The instruction says to speak to users using plain Chinese (“只说人话”), while the file also contains English description text and does not state that the skill is restricted to Chinese-speaking users only. This creates a language-policy issue because the skill appears to impose a language/locale expectation without user opt-in or explicit justification.

Static analysis

No suspicious patterns detected.