Setup
Interactive onboarding interview that builds the user's global memory profile. Use when user invokes /mematrix:setup, or asks to "set up my profile", "onboard me", "get to know me", "initialize my memories", or wants to establish their user identity/preferences for the first time.From its SKILL.md
npx -y skills add themarchrain/mematrix --skill setupAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
3 things to look at
- 27 days oldThe repository was created 27 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
5.2 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it
Mematrix Setup — Interactive Profile Interview
Build the user's global memory profile through a conversational interview. The agent actively drills the user with questions — the user does NOT fill in templates manually.
Core rules
- Ask ONE question at a time. This is a conversation, not a form.
- Drill deeper based on answers — follow up on interesting details.
- Don't rush. Cover each domain thoroughly before moving on.
- Write incrementally after each phase completes.
- If the user doesn't know or wants to skip, move on gracefully.
- Match the user's energy: verbose answers → explore deeper; terse answers → be concise.
Memory location
All memories go to ~/.mematrix/global/:
profile.md— identity, background, personalitytechnical.md— languages, frameworks, tools, environmentpreferences.md— work style, communication, editor configlife.md— optional life context (only if user provides meaningful content)
Phase 1: Profile → ~/.mematrix/global/profile.md
Opening question:
让我们从最基础的开始吧!你希望我怎么称呼你?另外,简单介绍一下你自己——比如你目前的工作、角色,或者你在做什么事情?
Drill-down topics (ask based on what they share, don't checklist all at once):
- Role & industry — "你目前具体做什么工作?在哪个行业?"
- Experience level — "你做这行多久了?之前还做过什么?"
- Education — "方便说一下你的教育背景吗?专业方向?"
- Personality & communication style — "你觉得自己是什么样性格的人?沟通上喜欢直截了当,还是娓娓道来?"
- Current focus — "当前主要精力花在什么事上?工作、学习、还是个人项目?"
- Goals — "近期有什么目标或者想达成的事情吗?"
- How they use AI — "日常用 AI 主要做什么?写代码、写作、学习、生活助手?"
Write to profile.md once this phase is complete.
Phase 2: Technical Stack → ~/.mematrix/global/technical.md
Opening question:
接下来聊聊技术方面!你主要用哪些编程语言?每种熟练程度大概怎么样?
Drill-down topics:
- Languages & proficiency — "除了这些,还有没有偶尔会用到的语言?"
- Frameworks — "主要用哪些框架?后端用 Spring Boot 还是 Gin?前端用 Vue 还是 React?"
- Tools & environment — "开发环境什么样?操作系统、编辑器、终端?"
- Package management — "包管理工具偏好?npm、pnpm、yarn?"
- Database — "数据库主要用什么?MySQL、PostgreSQL?"
- Infrastructure — "会用到 Docker、K8s、云服务吗?用的是哪些云?"
- Learning goals — "有没有在学的新技术或者接下来想学的?"
Write to technical.md once this phase is complete.
Phase 3: Preferences → ~/.mematrix/global/preferences.md
Opening question:
再说说你的工作习惯和偏好!写代码的时候有没有什么特别在意的——代码风格、注释习惯、测试方式这些?
Drill-down topics:
- Code style — "有没有具体的格式化偏好?缩进、引号、分号这些?"
- Testing — "写测试吗?习惯 TDD 还是先写代码再补测试?"
- Communication detail level — "和 AI 对话时,希望回答简洁一点,还是详细深入?"
- Language in code — "代码里注释和变量命名用中文还是英文?"
- Decision style — "做技术决策时,倾向于自己拿主意,还是希望 AI 给出多种方案对比?"
- Editor config — "编辑器有没有特别的配置习惯?主题、字体、快捷键?"
- Work schedule — "工作和作息大概什么样?什么时候最活跃?"
Write to preferences.md once this phase is complete.
Phase 4: Life Context (optional) → ~/.mematrix/global/life.md
Opening question:
最后,有没有一些生活相关的信息想让我记住的?比如在哪个城市、有什么兴趣爱好、或者其他你觉得 AI 应该了解的事情?
这部分是可选的,不想说完全可以跳过。
Topics (if user engages): location & timezone, languages spoken, hobbies, family context (if they bring it up), any constraints or preferences.
Only write to life.md if the user provides meaningful content.
During the interview
- Be conversational. Acknowledge answers naturally before the next question: "了解了,你在做金融科技的后端开发,主要用 Go 和 Java..."
- Drill, don't checklist. User mentions Go microservices → ask about framework.
- Allow redo. User can say "重新回答这一题" or "修改刚才那个".
After the interview
- Summarize what was captured across all global files
- Remind the user they can update via
/mematrix:rememberor redo specific sections with/mematrix:setup - Tell them: "从现在开始,用 /mematrix:recall 就会加载这些信息"
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.