CtrlK
BlogDocsLog inGet started
Tessl Logo

add-sample-data

This skill should be used when the user asks to "add sample data", "populate tables", "seed data", "add test records", "generate sample records", "insert demo data", "fill tables with data", "create test data", or wants to populate their Dataverse tables with sample records so they can test and demo their Power Pages site.

Install with Tessl CLI

npx tessl i github:microsoft/power-platform-skills --skill add-sample-data
What are skills?

88

Quality

87%

Does it follow best practices?

Impact

Pending

No eval scenarios have been run

SKILL.md
Review
Evals

Add Sample Data

Populate Dataverse tables with sample records via OData API so users can test and demo their Power Pages sites.

Core Principles

  • Respect insertion order: Always insert parent/referenced tables before child/referencing tables so lookup IDs are available when needed.
  • Use TaskCreate/TaskUpdate: Track all progress throughout all phases -- create the todo list upfront with all phases before starting any work.
  • Fail gracefully: On insertion failure, log the error and continue with remaining records -- never attempt automated rollback.

Initial request: $ARGUMENTS


Phase 1: Verify Prerequisites

Goal: Confirm PAC CLI auth, acquire an Azure CLI token, and verify API access

Actions:

  1. Create todo list with all 6 phases (see Progress Tracking table)
  2. Follow the prerequisite steps in ${CLAUDE_PLUGIN_ROOT}/references/dataverse-prerequisites.md to verify PAC CLI auth, acquire an Azure CLI token, and confirm API access. Store the environment URL as $envUrl.

Output: Authenticated session with valid token and confirmed API access


Phase 2: Discover Tables

Goal: Find the custom tables available in the user's Dataverse environment

Actions:

Path A: Read .datamodel-manifest.json (Preferred)

Check if .datamodel-manifest.json exists in the project root (written by the setup-datamodel skill). If it exists, read it -- it already contains table logical names, display names, and column info.

# Check if manifest exists
Test-Path ".datamodel-manifest.json"

See ${CLAUDE_PLUGIN_ROOT}/references/datamodel-manifest-schema.md for the full manifest schema.

Path B: Query OData API (Fallback)

If no manifest exists, discover custom tables via OData:

$tables = Invoke-RestMethod -Uri "$envUrl/api/data/v9.2/EntityDefinitions?`$select=LogicalName,DisplayName,EntitySetName&`$filter=IsCustomEntity eq true" -Headers $headers

For each discovered table, fetch its custom columns:

$columns = Invoke-RestMethod -Uri "$envUrl/api/data/v9.2/EntityDefinitions(LogicalName='<table>')/Attributes?`$select=LogicalName,DisplayName,AttributeType,RequiredLevel&`$filter=IsCustomAttribute eq true" -Headers $headers

2.1 Present Available Tables

Show the user the list of discovered tables with their columns so they can choose which to populate.

Output: List of discovered tables with their columns presented to the user


Phase 3: Select Tables & Configure

Goal: Gather user preferences on which tables to populate and how many records to create

Actions:

3.1 Select Tables

Use AskUserQuestion to ask which tables they want to populate (use multiSelect: true). List all discovered tables as options.

3.2 Select Record Count

Use AskUserQuestion to ask how many sample records per table:

OptionDescription
5 recordsQuick test -- just enough to verify the setup
10 recordsLight demo data for basic testing
25 recordsFuller dataset for realistic demos
CustomLet the user specify a number

3.3 Determine Insertion Order

Analyze relationships between selected tables. Parent/referenced tables must be inserted first so their IDs are available for child/referencing table lookups.

Build the insertion order:

  1. Tables with no lookup dependencies (parent tables) -- insert first
  2. Tables that reference already-inserted tables -- insert next
  3. Continue until all tables are ordered

Output: Confirmed table selection, record count, and insertion order


Phase 4: Generate & Review Sample Data

Goal: Generate contextually appropriate sample records and get user approval before inserting

Actions:

4.1 Generate Contextual Sample Data

For each selected table, generate sample records with contextually appropriate values based on column names and types:

  • String columns: Generate realistic values matching the column name (e.g., "Email" -> jane.doe@example.com, "Phone" -> (555) 123-4567, "Name" -> realistic names)
  • Memo columns: Generate short descriptive text relevant to the column name
  • Integer/Decimal/Currency columns: Generate reasonable numeric values
  • DateTime columns: Generate dates within a sensible range (past year to next month)
  • Boolean columns: Mix of true and false values
  • Picklist/Choice columns: Query valid options first (see references/odata-record-patterns.md), then use actual option values
  • Lookup columns: Reference records from parent tables that will be/were already inserted

4.2 Present Sample Data Preview

For each table, show a markdown table previewing the sample records directly in the conversation:

### Project (cr123_project) -- 5 records

| Name | Description | Status | Start Date |
|------|-------------|--------|------------|
| Website Redesign | Modernize the corporate website | 100000000 (Active) | 2025-03-15 |
| Mobile App | Build iOS and Android app | 100000001 (Planning) | 2025-04-01 |
| ... | ... | ... | ... |

Show relationship handling: which lookup fields reference which parent table records.

Output: Sample data plan ready for insertion. Proceed directly to Phase 5.


Phase 5: Insert Sample Data

Goal: Execute OData POST calls to create all approved sample records with correct relationship handling

Actions:

Refer to references/odata-record-patterns.md for full patterns.

5.1 Get Entity Set Names

For each table, get the entity set name (needed for the API URL):

$entityDef = Invoke-RestMethod -Uri "$envUrl/api/data/v9.2/EntityDefinitions(LogicalName='<table>')?`$select=EntitySetName" -Headers $headers
$entitySetName = $entityDef.EntitySetName

5.2 Get Picklist Options

For any picklist/choice columns, query valid option values before insertion:

$picklistMeta = Invoke-RestMethod -Uri "$envUrl/api/data/v9.2/EntityDefinitions(LogicalName='<table>')/Attributes(LogicalName='<column>')/Microsoft.Dynamics.CRM.PicklistAttributeMetadata?`$expand=OptionSet" -Headers $headers

Use the actual Value integers from the option set in your sample data.

5.3 Insert Parent Tables First

Insert records into parent/referenced tables first to capture their IDs:

$body = @{
    cr123_name = "Sample Record"
    cr123_description = "A sample record for testing"
} | ConvertTo-Json

$response = Invoke-RestMethod -Method Post -Uri "$envUrl/api/data/v9.2/<EntitySetName>" -Headers $headers -Body $body -ContentType "application/json"

Capture the returned record ID from the OData-EntityId response header or by querying back:

# The ID is in the response headers
# Or query: GET {envUrl}/api/data/v9.2/<EntitySetName>?$filter=cr123_name eq 'Sample Record'&$select=cr123_<table>id

Store parent record IDs for use in child table lookups.

5.4 Insert Child Tables with Lookups

For child/referencing tables, use @odata.bind syntax to set lookup fields:

$body = @{
    cr123_name = "Child Record"
    "cr123_ParentId@odata.bind" = "/<ParentEntitySetName>(<parent_guid>)"
} | ConvertTo-Json

Invoke-RestMethod -Method Post -Uri "$envUrl/api/data/v9.2/<ChildEntitySetName>" -Headers $headers -Body $body -ContentType "application/json"

5.5 Track Progress

Track each insertion attempt:

  • Record table name, record number, success/failure
  • On failure, log the error message but continue with remaining records
  • Do NOT attempt automated rollback on failure

5.6 Refresh Token Periodically

Refresh the Azure CLI token every 20 records to avoid expiration:

$token = az account get-access-token --resource "$envUrl" --query accessToken -o tsv
$headers["Authorization"] = "Bearer $token"

Output: All approved records inserted with parent-child relationships established


Phase 6: Verify & Summarize

Goal: Confirm record counts and present a final summary to the user

Actions:

6.1 Verify Record Counts

For each table that was populated, query the record count:

$count = Invoke-RestMethod -Uri "$envUrl/api/data/v9.2/<EntitySetName>?`$count=true&`$top=0" -Headers $headers

The @odata.count field in the response gives the total record count.

6.2 Record Skill Usage

Reference: ${CLAUDE_PLUGIN_ROOT}/references/skill-tracking-reference.md

Follow the skill tracking instructions in the reference to record this skill's usage. Use --skillName "AddSampleData".

6.3 Present Summary

Present a summary table:

TableRecords RequestedRecords CreatedFailures
cr123_project (Project)10100
cr123_task (Task)1091

Include:

  • Total records created across all tables
  • Any failures with error details
  • Lookup relationships that were established

6.4 Suggest Next Steps

After the summary, suggest:

  • Review the data in the Power Pages maker portal or model-driven app
  • If the site is not yet built: /power-pages:create-site
  • If the site is ready to deploy: /power-pages:deploy-site

Output: Verified record counts and summary presented to the user


Important Notes

Throughout All Phases

  • Use TaskCreate/TaskUpdate to track progress at every phase
  • Ask for user confirmation at key decision points (see list below)
  • Respect insertion order -- always insert parent tables before child tables
  • Fail gracefully -- log errors and continue, never rollback automatically
  • Refresh tokens every 20 records to avoid expiration

Key Decision Points (Wait for User)

  1. After Phase 2: Confirm which tables to populate
  2. After Phase 3: Confirm record count and insertion order
  3. After Phase 6: Review summary and decide next steps

Progress Tracking

Before starting Phase 1, create a task list with all phases using TaskCreate:

Task subjectactiveFormDescription
Verify prerequisitesVerifying prerequisitesConfirm PAC CLI auth, acquire Azure CLI token, verify API access
Discover tablesDiscovering tablesRead .datamodel-manifest.json or query OData API for custom tables
Select tables and configureConfiguring tablesUser picks tables, record count, and determine insertion order
Generate and review sample dataGenerating sample dataGenerate contextual sample records, present preview, get user approval
Insert sample dataInserting recordsExecute OData POST calls with relationship handling and token refresh
Verify and summarizeVerifying resultsConfirm record counts, present summary, suggest next steps

Mark each task in_progress when starting it and completed when done via TaskUpdate. This gives the user visibility into progress and keeps the workflow deterministic.


Begin with Phase 1: Verify Prerequisites

Repository
microsoft/power-platform-skills
Last updated
Created

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.