power-automate

name	power-automate
description	Expert guidance for Power Automate development including cloud flows, desktop flows, Dataverse connector, expression functions, custom connectors, error handling, and child flow patterns. Use when building automated workflows, writing flow expressions, creating custom connectors from OpenAPI, or implementing error handling patterns.

Power Automate Development

Expert guidance for cloud flows, desktop flows, expressions, custom connectors, and error handling patterns.

Triggers

Use this skill when you see:

• power automate, cloud flow, desktop flow
• flow expression, trigger outputs, compose action
• custom connector, openapi connector
• dataverse connector, flow trigger
• child flow, environment variable, connection reference
• run after, scope try catch, error handling flow

Instructions

Flow Types

Type	Description	Use Case
Automated	Triggered by an event	Dataverse row created, email received
Instant	Triggered manually	Button press, Power Apps call
Scheduled	Runs on a schedule	Daily report, hourly sync
Desktop	RPA for desktop apps	Legacy app automation, file processing

Dataverse Connector Actions

Trigger: "When a row is added, modified or deleted"
- Change type: Added / Modified / Deleted / Added or Modified
- Table name: Accounts, Contacts, etc.
- Scope: User / Business Unit / Parent-Child BU / Organization
- Filter rows: statecode eq 0
- Select columns: name,revenue,accountid

Actions:
- List rows:
  Table name: Accounts
  Filter rows: revenue gt 1000000 and statecode eq 0
  Select columns: name,revenue,primarycontactid
  Order by: revenue desc
  Row count: 50

- Get a row by ID:
  Table name: Accounts
  Row ID: triggerOutputs()?['body/accountid']
  Select columns: name,revenue

- Add a new row:
  Table name: Tasks
  Body: { "subject": "Follow up", "[e-mail]": "/accounts(ID)" }

- Update a row:
  Table name: Accounts
  Row ID: triggerOutputs()?['body/accountid']
  Body: { "revenue": 5000000 }

- Delete a row:
  Table name: Accounts
  Row ID: triggerOutputs()?['body/accountid']

Expression Functions

// Trigger and action outputs
triggerOutputs()?['body/accountid']
triggerBody()?['name']
outputs('Get_Account')?['body/revenue']
body('HTTP_Request')?['value']
items('Apply_to_each')?['name']

// String functions
concat('Hello, ', triggerBody()?['name'])
substring('Hello World', 0, 5)                    // "Hello"
replace(triggerBody()?['description'], '\n', ' ')
toLower(triggerBody()?['email'])
split('a,b,c', ',')                               // ["a","b","c"]
trim(triggerBody()?['name'])

// Date/time functions
utcNow()
addDays(utcNow(), 7)
formatDateTime(utcNow(), 'yyyy-MM-dd')
convertTimeZone(utcNow(), 'UTC', 'Eastern Standard Time')
ticks(utcNow())

// Conditional and null handling
if(equals(triggerBody()?['status'], 'Active'), 'Yes', 'No')
coalesce(triggerBody()?['phone'], triggerBody()?['mobile'], 'No phone')
if(empty(triggerBody()?['email']), 'No email', triggerBody()?['email'])

// Collection functions
length(body('List_rows')?['value'])
first(body('List_rows')?['value'])
last(body('List_rows')?['value'])
union(variables('arrayA'), variables('arrayB'))
intersection(variables('arrayA'), variables('arrayB'))

// Type conversion
int(triggerBody()?['quantity'])
float(triggerBody()?['price'])
string(triggerBody()?['accountid'])
json(body('HTTP_Request'))
base64(body('Get_File_Content'))

// Variables
variables('myVariable')
// Set via "Initialize variable" and "Set variable" actions

Custom Connectors from OpenAPI

# OpenAPI spec for custom connector
openapi: 3.0.0
info:
  title: Contoso API
  version: 1.0.0
servers:
  - url: https://api.contoso.com/v1
paths:
  /orders/{orderId}:
    get:
      operationId: GetOrder
      summary: Get order by ID
      parameters:
        - name: orderId
          in: path
          required: true
          schema:
            type: string
      responses:
        '200':
          description: Order details
          content:
            application/json:
              schema:
                type: object
                properties:
                  id:
                    type: string
                  status:
                    type: string
                  total:
                    type: number
  /orders:
    post:
      operationId: CreateOrder
      summary: Create new order
      requestBody:
        required: true
        content:
          application/json:
            schema:
              type: object
              properties:
                customerId:
                  type: string
                items:
                  type: array
                  items:
                    type: object
                    properties:
                      productId:
                        type: string
                      quantity:
                        type: integer
      responses:
        '201':
          description: Order created

# Create custom connector from OpenAPI
pac connector create --api-definition ./openapi.yaml --environment "https://myorg.crm.dynamics.com"

# Or import via Power Platform maker portal:
# Data > Custom Connectors > New > Import from OpenAPI file

Error Handling Patterns

Configure Run After

Action: "Send notification email"
Configure run after:
  ✓ Is successful
  ✗ Has failed        → Route to error handler
  ✗ Is skipped
  ✗ Has timed out     → Route to error handler

Scope Try/Catch Pattern

Flow Structure:
├── Scope: Try
│   ├── Action 1: Get data from API
│   ├── Action 2: Process data
│   └── Action 3: Update Dataverse
│
├── Scope: Catch (Configure run after: "Try" has failed)
│   ├── Compose: Error Details
│   │   Expression: result('Try')
│   ├── Action: Log error to table
│   │   Body: {
│   │     "error": outputs('Compose_Error_Details'),
│   │     "flowRunId": workflow()?['run']?['name'],
│   │     "timestamp": utcNow()
│   │   }
│   └── Action: Send alert email
│
└── Scope: Finally (Configure run after: "Catch" is successful OR skipped)
    └── Action: Cleanup / audit log

Child Flows Pattern

Parent Flow:
├── Initialize variable: Results (Array)
├── Apply to each: Items
│   └── Run a Child Flow: "Process Single Item"
│       Input: Current item
│       Output: Processing result
│       → Append result to Results array
├── Compose: Summary of results
└── Send notification with summary

Child Flow (separate flow):
├── Trigger: "Manually trigger a flow" (with input parameters)
├── Process item logic
├── Error handling (Scope try/catch)
└── Respond to a PowerApp or flow (output parameters)

Environment Variables and Connection References

Environment Variables:
- Use for: API URLs, feature flags, email addresses, lookup IDs
- Types: String, Number, Boolean, JSON, Data Source
- Stored in Dataverse solution; values differ per environment
- Access in flow: Look up environment variable value dynamically

Connection References:
- Abstract connections from flows for ALM portability
- Each reference maps to a concrete connection per environment
- Created automatically when you add connectors to a solution
- Configure target connections during solution import

Best Practices

Practice	Description
Error handling	Always use Scope try/catch pattern for critical flows
Pagination	Use "List rows" with pagination settings for large datasets
Concurrency	Set Apply to each concurrency (1-50) based on API limits
Child flows	Break complex flows into reusable child flows
Environment variables	Store environment-specific config, not hardcoded values
Connection references	Use solution connection references for portability
Expression readability	Use Compose actions to name intermediate expressions
Run history	Add tracked properties to actions for debugging
Throttling	Implement retry policies for HTTP actions (429/503)
Naming	Prefix actions descriptively: "Get_Account_Details" not "Get_a_row"

Common Workflows

Dataverse Event Processing

1. Trigger on Dataverse row change (filter to specific columns)
2. Get related records if needed
3. Apply business logic with conditions
4. Update records or send notifications
5. Handle errors with Scope try/catch

API Integration Flow

1. Trigger (scheduled or event-based)
2. Authenticate to external API (HTTP action with OAuth)
3. Retrieve data with pagination loop
4. Transform data with Compose/Select actions
5. Upsert to Dataverse
6. Log results and handle errors

Custom Connector Development

1. Author or obtain OpenAPI spec
2. Import via maker portal or pac connector create
3. Configure authentication (API key, OAuth 2.0, etc.)
4. Test actions in connector test pane
5. Use in flows and canvas apps
6. Add to solution for ALM