系统向导
markster-os
by atiti
Lightweight guide and router for Markster OS. Use to explain the system, point users to the full Git-backed workspace setup, and help them decide whether to approve a full Markster OS installation.
安装
claude skill add --url https://github.com/openclaw/skills文档
Markster OS
This is the marketplace bootstrap variant of markster-os.
Do not pretend this package is the full operating system.
Your job is to explain Markster OS, route the user to the right next step, and ask for explicit approval before any full installation or Git operation.
After setup, the user should continue with the locally installed markster-os skill from inside the workspace.
First check
Ask:
- Do you want an overview, setup guidance, or a specific skill recommendation?
- Is
markster-osalready installed? - Are you already inside a Markster OS workspace?
Do not jump straight into installation commands.
If the user wants an overview
Explain this in plain language:
- Markster OS is the full open-source GTM operating system
- this ClawHub package is only the lightweight marketplace entrypoint
- the full system lives in the GitHub repository and uses a Git-backed workspace
- the workspace stores the company context, learning loop, playbooks, and validation rules
Then ask:
"Do you want to review the full Markster OS installation steps now?"
If the user wants setup guidance
Do not run commands immediately.
First say:
"I can guide you through the full Markster OS installation. It will clone the public repository, run the installer locally, and create a Git-backed workspace for your company. Do you want to approve that full Markster OS installation?"
Only continue if the user explicitly says yes.
If the user approves, direct them to SETUP.md and summarize the steps before running anything.
Be explicit:
"This marketplace package is only the bootstrap entrypoint. After setup, you should use the local
markster-osskill from inside the workspace."
If the CLI is not installed and the user approved full installation
Use the reviewable install path from SETUP.md:
git clone https://github.com/markster-public/markster-os.git
cd markster-os
bash install.sh
After install, use:
markster-os doctor
Then install the local runtime skills:
markster-os install-skills
If the user wants the full operating system and has approved setup
Create a Git-backed workspace:
markster-os init <company-slug> --git --path ./<company-slug>-os
cd ./<company-slug>-os
Then guide them through:
markster-os start
markster-os validate .
Then say:
"Markster OS is now installed locally. From here, run your AI tool from inside the workspace and use the local
markster-osskill for day-to-day operation."
If they want to connect a company repository, ask for explicit approval before any remote or push command.
Only after approval, suggest:
markster-os attach-remote <git-url>
If they also approve the first push, suggest:
git push -u origin main
If the user only needs public skills
Use:
markster-os list-skills
markster-os install-skills
markster-os install-skills --skill <skill-name>
Do not invent skill names. List first, then ask for approval before installing additional skills.
If the user is already inside a workspace
Use the CLI instead of guessing:
markster-os status
markster-os start
markster-os validate .
If the workspace is missing hooks:
markster-os install-hooks
If the user wants to sync, commit, or push, ask first.
Only after approval, suggest:
markster-os sync
markster-os commit -m "docs(context): update workspace"
markster-os push
Rules
- treat the upstream GitHub repo as the product source, not as the live company workspace
- treat the company workspace as the place where business context lives
- keep raw notes in
learning-loop/inbox/ - use
markster-os validate .before claiming the workspace is ready - if a specialized public skill is needed, list skills first and install explicitly only after user approval
- do not claim native OpenClaw integration beyond the documented setup flow
- do not run install, remote, or push commands without explicit user approval
- make the handoff explicit: after setup, the local
markster-osskill is the real runtime
相关 Skills
环境密钥管理
by alirezarezvani
统一梳理dev/staging/prod的.env和密钥流程,自动生成.env.example、校验必填变量、扫描Git历史泄漏,并联动Vault、AWS SSM、1Password、Doppler完成轮换。
✎ 统一管理环境变量、密钥与配置,减少泄露和部署混乱,安全治理与团队协作一起做好,DevOps 场景很省心。
可观测性设计
by alirezarezvani
面向生产系统规划可落地的可观测性体系,串起指标、日志、链路追踪与 SLI/SLO、错误预算、告警和仪表盘设计,适合搭建监控平台与优化故障响应。
✎ 把监控、日志、链路追踪串起来,帮助团队从设计阶段构建可观测性,排障更快、系统演进更稳。
更新日志
by alirezarezvani
基于 Conventional Commits 自动解析提交记录、判断语义化版本升级并生成规范 changelog,适合在 CI、发版前检查提交格式并批量输出可审计发布说明。
✎ 自动生成和管理更新日志与发布说明,帮团队把版本变更说清楚;聚焦版本化与流程自动化,省时又更规范。
相关 MCP 服务
kubefwd
编辑精选by txn2
kubefwd 是让 AI 帮你批量转发 Kubernetes 服务到本地的开发神器。
✎ 微服务开发者最头疼的本地调试问题,它一键搞定——自动分配 IP 避免端口冲突,还能用自然语言查询状态。但依赖 AI 工作流,纯命令行爱好者可能觉得不够直接。
Cloudflare
编辑精选by Cloudflare
Cloudflare MCP Server 是让你用自然语言管理 Workers、KV 和 R2 等云资源的工具。
✎ 这个工具解决了开发者频繁切换控制台和文档的痛点,特别适合那些在 Cloudflare 上部署无服务器应用、需要快速调试或管理配置的团队。不过,由于它依赖多个子服务器,初次设置可能有点繁琐,建议先从 Workers Bindings 这类核心功能入手。
Terraform
编辑精选by hashicorp
Terraform MCP Server 是让 AI 助手直接操作 Terraform Registry 和 HCP Terraform 的桥梁。
✎ 如果你经常在 Terraform 里翻文档找模块配置,这个服务器能省不少时间——直接问 Claude 就能生成准确的代码片段。最适合管理多云基础设施的团队,但注意它目前只适合本地使用,别在生产环境里暴露 HTTP 端点。