Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed assistant for designing OpenClaw skill packages and does not show hidden installation, persistence, credential access, exfiltration, or destructive behavior.
Install this if you want an agent to help design OpenClaw skills. Be aware it may create generated files in an output directory after you confirm the design, and review any generated skill before installing it into an agent.
The trigger conditions are broad phrases like '创建 skill', '做一个技能', and '加一个能力', which can appear in ordinary discussion and cause accidental activation. Unintended invocation is risky here because the skill is designed to generate installation-ready artifacts and may steer the conversation into producing files or instructions the user did not intend to request.
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
description: 通过需求收口帮用户从零设计一个 OpenClaw agent skill。逐步收集目标、输入、输出、过程控制，确认后输出完整的 skill 文件包（SKILL.md + 安装指南）。产出是文件，不替用户执行安装。当用户说"创建 skill"、"做一个技能"、"加一个能力"、"帮我把 XXX 变成 skill"时触发。
---
# Create Skill
通过需求收口帮用户设计 skill，输出完整可安装的 skill 文件包。
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.
description: 通过需求收口帮用户从零设计一个 OpenClaw agent skill。逐步收集目标、输入、输出、过程控制,确认后输出完整的 skill 文件包(SKILL.md + 安装指南)。产出是文件,不替用户执行安装。当用户说"创建 skill"、"做一个技能"、"加一个能力"、"帮我把 XXX 变成 skill"时触发。
---
# Create Skill
通过需求收口帮用户设计 skill,输出完整可安装的 skill 文件包。
The skill explicitly claims it will only output a file package and will not modify the environment, but later instructs writing generated files into a local working directory. That creates a mismatch between declared behavior and actual behavior, which can lead the hosting agent to perform filesystem writes when the user expected a purely advisory response.
No suspicious patterns detected.