ClawHub
Back to skill

Security audit

詹明明·拿不准的时候

Security checks for vulnerabilities and agentic risk

Overview

This is a decision-support skill with disclosed, purpose-aligned memory use and no executable code, network retrieval, install steps, or hidden destructive behavior.

Before installing, be aware that ordinary phrases like being stuck or unsure may invoke this skill, and that it records decision assumptions and follow-up notes in configured zmm memory. Use it when you are comfortable having business decisions summarized for later review.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
95% confidence
Finding
The trigger list contains broad everyday phrases such as '要不要做', '卡住了', and '想不清楚' that can easily appear in normal conversation outside the intended decision-support context. This can cause unintended invocation, routing users into the skill when they did not explicitly request it, increasing the chance of irrelevant guidance, context hijacking, or accidental memory reads/writes tied to this skill.

Natural-Language Policy Violations

Medium
Confidence
88% confidence
Finding
The skill instructs the agent to '只说人话' in Chinese-centric guidance and is designed primarily around Chinese interaction without an explicit mechanism to detect and honor the user's language preference. While not a classic security flaw, this can create unsafe misunderstandings if the user operates in another language and receives inaccurate or partially understood decision guidance.

Static analysis

No suspicious patterns detected.