Token导航 LogoToken导航TokenDH.com
运维和基础设施external-servicegithub未标认证来源可访问clear审计通过

powerplatformpowerplatform 命令行

Agent Skill

powerplatform 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

941

周安装

40

GitHub Stars

1

下载量

330
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

3

许可证

MIT

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:powerplatform(powerplatform 命令行)
来源仓库:https://github.com/robdtaylor/personal-ai-infrastructure
仓库路径:skills/powerplatform
安装命令:
npx skills add https://github.com/robdtaylor/personal-ai-infrastructure --skill Powerplatform
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/robdtaylor/personal-ai-infrastructure --skill Powerplatform

简介

powerplatform 用于处理 GitHub 仓库、Issue、Pull Request 等协作信息,适合在代码协作场景中整理变更与状态。

  • 适用于围绕仓库状态、代码变更或协作事项的信息组织与汇总。
  • 通过 npx skills add 命令从 GitHub 仓库安装并使用。
  • 安装前应确认权限范围、维护状态及是否执行命令或访问文件。
  • 具体功能需参考原始 README 进一步验证。

SKILL.md

PowerPlatform - Flow Generation Skill

Generate Power Automate flows from natural language descriptions using connector schemas exported from your work environment.

Examples

Example: Simple notification flow

Example: Approval workflow

User: "Make a flow for document approval - when a file is added to SharePoint, start an approval, then move to approved/rejected folder"
-> Uses SharePoint trigger + Approvals + SharePoint actions
-> Includes conditional branching
-> Outputs deployable solution

Architecture

Work Repo (schemas)                    PAI (this skill)
~/Projects/work/scripts/               ~/.claude/skills/PowerPlatform/
power-platform/
├── connectors/          ──reads──>    ├── SKILL.md
│   ├── shared_teams.json              ├── templates/
│   ├── shared_outlook.json            ├── reference/
│   └── ...                            └── examples/
├── manifest.json
└── generated/           <──writes──   [Solution ZIPs]

Workflow

1. Check Available Connectors

Before generating, read the manifest to see what connectors are available:

cat ~/Projects/work/scripts/power-platform/manifest.json

If manifest.json doesn't exist or is stale, inform user to run the export script at work.

2. Parse User Request

Extract from natural language:

  • Trigger: What starts the flow (email arrives, file created, scheduled, manual)
  • Actions: What the flow does (send message, create item, update record)
  • Conditions: Any branching logic (if/then, switch)
  • Data flow: What data passes between steps

3. Generate Flow Definition

Create the flow JSON following Power Automate schema. Key structure:

{
  "properties": {
    "definition": {
      "$schema": "https://schema.management.azure.com/providers/Microsoft.Logic/schemas/2016-06-01/workflowdefinition.json#",
      "triggers": { ... },
      "actions": { ... }
    },
    "connectionReferences": { ... }
  }
}

4. Package as Solution

Wrap in solution structure for import:

solution/
├── [Content_Types].xml
├── customizations.xml
├── solution.xml
└── Workflows/
    └── {flow-guid}-{flow-name}.json

5. Output

Save to: ~/Projects/work/scripts/power-platform/generated/


Connection References

Flows use connection references that map to actual connections in the target environment. Use placeholder IDs that will be resolved on import:

"connectionReferences": {
  "shared_teams": {
    "connectionName": "shared_teams",
    "source": "Embedded",
    "id": "/providers/Microsoft.PowerApps/apis/shared_teams",
    "tier": "NotSpecified"
  }
}

On import, Power Platform prompts user to map these to their actual connections.


Expression Syntax

Power Automate uses a specific expression language. Common patterns:

Dynamic Content References

@{triggerOutputs()?['body/subject']}           # Email subject from trigger
@{body('Get_item')?['Title']}                  # Field from previous action
@{items('Apply_to_each')?['name']}             # Current item in loop

Functions

@{utcNow()}                                    # Current timestamp
@{concat('Hello ', triggerBody()?['name'])}   # String concatenation
@{if(equals(1,1),'yes','no')}                 # Conditional
@{length(body('Get_items')?['value'])}        # Array length
@{formatDateTime(utcNow(),'yyyy-MM-dd')}      # Date formatting

Conditions

"expression": {
  "and": [
    { "contains": ["@triggerOutputs()?['body/subject']", "urgent"] },
    { "equals": ["@triggerOutputs()?['body/importance']", "high"] }
  ]
}

See reference/expressions.md for full syntax guide.


Common Triggers

TriggerConnectorUse Case
When_a_new_email_arrivesOffice365Email automation
When_a_file_is_createdSharePointDocument workflows
When_an_item_is_createdSharePointList automation
When_a_row_is_addedDataverseDatabase triggers
RecurrenceScheduleScheduled jobs
manualManualButton-triggered
When_a_HTTP_request_is_receivedHTTPWebhook/API trigger

Common Actions

ActionConnectorUse Case
Send_an_emailOffice365Email notifications
Post_message_in_a_chat_or_channelTeamsTeam notifications
Create_itemSharePointAdd list items
Update_itemSharePointModify list items
Start_and_wait_for_an_approvalApprovalsApproval workflows
Create_a_rowDataverseDatabase writes
HTTPHTTPExternal API calls

Templates

Pre-built patterns in templates/:

  • approval-flow.json - Standard approval workflow
  • notification-flow.json - Event-triggered notifications
  • scheduled-report.json - Scheduled data collection
  • form-processing.json - Forms response handling
  • file-sync.json - Cross-system file operations

Generation Process

When asked to create a flow:

  1. Read available schemas: ls ~/Projects/work/scripts/power-platform/connectors/ cat ~/Projects/work/scripts/power-platform/manifest.json
  2. Load relevant connector schemas for the requested functionality
  3. Select appropriate template or build from scratch
  4. Generate flow JSON with:

- Unique GUID for flow - Proper trigger configuration - Action sequence with correct operation IDs - Connection references for all connectors used - Expression syntax for dynamic content

  1. Create solution package using tools/package-solution.sh
  2. Save to generated folder and inform user

Limitations

  • Cannot deploy directly - User must import at work
  • Connection mapping required - User maps connections on import
  • Premium connectors - Some require premium licenses
  • Custom connectors - Not supported without schema export
  • Schema freshness - Depends on export frequency at work

Files Reference

FilePurpose
templates/*.jsonPre-built flow patterns
reference/expressions.mdExpression syntax guide
reference/triggers.mdTrigger configurations
reference/actions.mdAction configurations
examples/*.jsonWorking example flows
tools/package-solution.shSolution packager script

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

能力 5

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

Claude Code

28.34%
按下载量换算94

OpenCode

23.7%
按下载量换算78

Cursor

20.47%
按下载量换算68

Gemini CLI

13.6%
按下载量换算45

windsurf

8.96%
按下载量换算30

Codex

3.4%
按下载量换算11

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

external-service

该 Skill 可能调用第三方服务、云服务或外部模型 API,使用前需要确认账号、额度、数据发送范围和服务条款。

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。

来源信息

继续浏览同类 Skills