Token导航 LogoToken导航TokenDH.com
研究检索敏感数据github未标认证来源可访问许可证需确认审计通过

datahub-mfe-configure-appdatahub mfe 配置应用程序

Agent Skill

用于辅助数据整理、表格处理、CSV/Excel 分析、指标计算和图表准备。它适合让 Agent 清洗字段、汇总数据、发现异常、生成统计口径或把分析结果转成可读说明。使用时需要确认数据来源、字段含义和时间范围,避免把样本数据当全量事实;涉及敏感数据、导出文件或批量写回时,应先确认权限和脱敏边界。

总安装

823

周安装

35

GitHub Stars

19

下载量

288
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:datahub-mfe-configure-app(datahub mfe 配置应用程序)
来源仓库:https://github.com/datahub-project/datahub-skills
仓库路径:skills/datahub-mfe-configure-app
安装命令:
npx skills add https://github.com/datahub-project/datahub-skills --skill datahub-mfe-configure-app
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/datahub-project/datahub-skills --skill datahub-mfe-configure-app

简介

注册微前端应用到 DataHub 前端系统,配置本地开发和生产环境路由。

  • 指导 Webpack Module Federation 集成和侧边栏导航配置,支持 Kubernetes 部署。
  • 包含常见问题排查指南,解决 MFE 加载失败和样式冲突问题。
  • 修改核心配置文件前务必备份原始版本,防止配置错误导致服务不可用。
  • datahub-mfe-configure-app 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Configure an MFE in DataHub

Walks through registering a Micro Frontend app with the DataHub frontend so it loads at /mfe/<path> and optionally appears in the navigation sidebar. Covers local development, production/k8s deployment, and troubleshooting.

Step 1: Gather Information

Use the AskQuestion tool to collect the following in a single call.

QuestionIDOptions
Are you configuring for local development or production/k8s?env_targetLocal development / Production / Kubernetes
MFE app name — the name field inside new ModuleFederationPlugin({name: '...'}) in the MFE's webpack.config.js. Must match exactly.mf_nametestAppMFE / N/A - I'll fill it in later
URL path where the MFE should be accessible, e.g. /dashboard (must start with /)mfe_path/test_page / N/A - I'll fill it in later
remoteEntry.js URL, e.g. http://localhost:3002/remoteEntry.jsremote_entry_urlhttp://localhost:3002/remoteEntry.js / N/A - I'll fill it in later
Display label for navigation, e.g. Team Dashboarddisplay_labelTest Dashboard / N/A - I'll fill it in later
Should the MFE appear in the left nav sidebar?show_in_navYes / No
Nav iconnav_iconChartBar / Trophy / Gear / HandWaving / Lightning / MagnifyingGlass / Database / Users / Shield / Bell / Flag / Star / Heart / Cube / Table / Code / Globe / Rocket

If env_target is Local development, ask one follow-up question using AskQuestion before proceeding:

QuestionIDOptions
How are you running DataHub locally?local_modeDocker / datahub-dev.sh (standard) / Play server directly (sbt / IntelliJ)

For the N/A - I'll fill it in later fields (mf_name, mfe_path, remote_entry_url, display_label), do NOT invent or suggest values from the workspace. The user will type their own.

Finding the MFE app name: Open the MFE's webpack.config.js and look for the ModuleFederationPlugin block. The name property is what goes here: ``js new ModuleFederationPlugin({ name: 'teamDashboardMFE', ... }) // ^^^^^^^^^^^^^^^^^ ← this value ` If the name in the YAML config doesn't match this exactly, the MFE will fail to load. If you used the datahub-mfe-create-app` skill, the name was printed in the summary as "Module Federation name".

Step 2: Generate the YAML Config Entry

Build the config entry from the gathered values:

  - id: __MFE_ID__
    label: __DISPLAY_LABEL__
    path: __MFE_PATH__
    remoteEntry: __REMOTE_ENTRY_URL__
    module: __MF_NAME__/mount
    flags:
      enabled: true
      showInNav: __SHOW_IN_NAV__
    navIcon: __NAV_ICON__

Where __MFE_ID__ is derived from the path (strip leading /, replace / with -), e.g. path /dashboard becomes id dashboard.

Step 3: Apply the Configuration

Local Development — Docker / datahub-dev.sh (standard)

The Docker image has MFE_CONFIG_FILE_PATH baked in at build time pointing to /datahub-frontend/conf/mfe.config.dev.yaml (set in docker/datahub-frontend/Dockerfile). This is the file to edit for any Docker-based workflow.

  1. Read the current datahub-frontend/conf/mfe.config.dev.yaml.
  2. Append the new entry under microFrontends:. If the file is empty, use:
subNavigationMode: false
microFrontends:
  - id: ...
    ...
  1. Rebuild the frontend container so the updated file is baked into the image:
scripts/dev/datahub-dev.sh rebuild --wait

Local Development — Play server directly (sbt / IntelliJ)

datahub-frontend/run/frontend.env points MFE_CONFIG_FILE_PATH to ../conf/mfe.config.local.yaml. Edit that file instead.

  1. Read the current datahub-frontend/conf/mfe.config.local.yaml.
  2. Append the new entry under microFrontends:.
  3. Restart the Play server — no rebuild needed, the file is read from disk:
cd datahub-frontend/run && ./run-local-frontend

Production / Kubernetes

Tell the user:

Since you selected Production / Kubernetes, I won't edit any local config files — instead here's what you need.
  1. Show the user the complete YAML config to add to their production config file (which may live in a separate config repo).
  2. Explain the env var and volume mount setup. See reference.md for the full k8s ConfigMap pattern.
  3. Remind the user about MFE_PUBLIC_PATH:
When deploying your MFE to a CDN or static host, set the MFE_PUBLIC_PATH env var during the MFE's production build so webpack knows the correct public URL for chunk loading. ``bash MFE_PUBLIC_PATH=https://cdn.example.com/my-mfe/ npm run build ``

Step 4: Verify

Guide the user through these checks:

  1. Standalone check: Open __REMOTE_ENTRY_URL__ in a browser — should return JavaScript content (not 404 or HTML).
  2. Integration check: Navigate to http://localhost:9002/mfe__MFE_PATH__ (local) or the equivalent production URL. The MFE should render inside the DataHub chrome.
  3. Nav check (if showInNav: true): The sidebar should show the __DISPLAY_LABEL__ item with the chosen icon. Click it to navigate.

If any check fails, see the Troubleshooting section in reference.md.

Step 5: Summary

Provide the user with a summary. Show only the config file relevant to their env_target — do not mention local config files if they selected Production.

If local development:

MFE registered in DataHub:
  Path:            /mfe__MFE_PATH__
  Remote entry:    __REMOTE_ENTRY_URL__
  Module:          __MF_NAME__/mount
  Nav sidebar:     __SHOW_IN_NAV__ (icon: __NAV_ICON__)
  Config file:     datahub-frontend/conf/mfe.config.dev.yaml    (Docker / datahub-dev.sh)
                   datahub-frontend/conf/mfe.config.local.yaml  (Play server / IntelliJ)

If Production / Kubernetes:

MFE registered in DataHub:
  Path:            /mfe__MFE_PATH__
  Remote entry:    __REMOTE_ENTRY_URL__
  Module:          __MF_NAME__/mount
  Nav sidebar:     __SHOW_IN_NAV__ (icon: __NAV_ICON__)
  Config file:     <your production config file>

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

37.87%
按下载量换算109

Claude

30.23%
按下载量换算87

Cursor

18.02%
按下载量换算52

Gemini CLI

9.33%
按下载量换算27

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills