kais-horse
PassAudited by ClawScan on May 10, 2026.
Overview
Kai's Horse is a prompt-only philosophy/thought-experiment skill; it asks to stay active for related topics but shows no code execution, credential use, file access, networking, or data exfiltration.
This appears safe to install if you want a persistent philosophy/thought-experiment helper. Be aware that it is designed to stay active and affect related conversations automatically; otherwise, there are no artifacts showing executable code, network calls, credentials, or local data access.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The skill may shape the agent's style and content in related conversations even when you did not explicitly invoke it.
This asks the agent to keep the skill globally active and automatically apply its philosophy framework to related topics, not only when the user types /kh commands.
[Mode] type = global priority = 9 always_on = true ... 日常对话静默常驻，仅在哲学、意识、伦理、AI相关话题自动联动谱系
Install it only if you want a globally available philosophy/AI-ethics assistant; disable it if you prefer skills to run only on explicit commands.
The skill is intended to remain loaded and auto-activate for certain discussion topics.
The usage guide explicitly describes silent background behavior and automatic topic recognition. In this artifact set it is prompt-level behavior, with no code or OS-level persistence.
【额外特性】 1. 后台静默运行，不干扰日常聊天 2. 自动识别哲学、AI、伦理、意识类话题
Review the host application's always-on/global-skill settings and turn the skill off if this background conversational activation is not desired.