ClawHub
Back to skill

Security audit

S2 Hanzi Antonym Logic

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed style and reasoning overlay for environmental-control language, with no bundled code, tools, installs, persistence, or direct hardware authority.

Install this only if you want the agent to use stylized Hanzi-based language for ambience and spatial-design discussions. For real smart-home control, keep normal numeric settings and confirmations available, especially for temperature, brightness, noise, accessibility, or safety-sensitive cases.

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 (3)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The documented trigger phrases are overly broad and map to ordinary natural-language requests about comfort, ambiance, or environment. That increases the chance the skill activates during routine user interactions and overrides expected device-control semantics, which can cause unintended actions or confusing behavior in systems connected to real smart-home hardware.

Natural-Language Policy Violations

Medium
Confidence
81% confidence
Finding
The README instructs users to abandon ordinary commands and use a specific philosophical/register-based interaction style without offering an alternative. By itself this is mostly a UX issue, but in an agent skill it can obscure user intent, reduce predictability, and make it harder for users to understand when the skill is controlling behavior versus merely rephrasing output.

Natural-Language Policy Violations

Medium
Confidence
96% confidence
Finding
The skill explicitly instructs the agent to abandon standard numeric controls and to always use a Hanzi-based dialectical interaction model, without checking user preference, accessibility needs, or task suitability. This can degrade clarity and operational safety in environment-control contexts because users may need precise values, unambiguous status, or plain-language confirmations, and the skill suppresses those in favor of stylized output.

Static analysis

No suspicious patterns detected.