Showing posts with label Managed Identity. Show all posts
Showing posts with label Managed Identity. Show all posts

Monday, 26 January 2026

React 19: The Practical Difference Between Hooks and Components (With TypeScript and Azure Integration)

The difference between react hooks and components matters because it defines how you separate logic from presentation. Problem: teams mix stateful logic directly inside UI and struggle to test, reuse, and scale. Solution: put data fetching, validation, and side effects in reusable hooks; keep rendering in lean components. Value: cleaner architecture, easier testing, fewer bugs, and production-ready integration with Azure using least privilege.

The Problem

Developers often blur the line between where logic lives (hooks) and where UI renders (components). This leads to duplicated code, tangled effects, and UI tests that are slow and brittle. We need a clear pattern: hooks encapsulate logic and I/O; components focus on layout and accessibility.

Prerequisites

Node.js v20+, TypeScript 5+ with strict mode, React 19, TanStack Query v5+, Zod v3+, Azure Functions Core Tools v4+, .NET 8 SDK, Azure CLI (or azd), and a browser-compatible fetch API.

The Solution (Step-by-Step)

1) Define strict runtime and compile-time types

// src/schemas/user.ts
import { z } from "zod";

// Zod schema for runtime validation and safe parsing
export const UserSchema = z.object({
  id: z.string().uuid(),
  email: z.string().email(),
  name: z.string().min(1),
});

export const UsersSchema = z.array(UserSchema);

export type User = z.infer<typeof UserSchema>;

2) Create a focused hook: logic, data fetching, and validation

// src/hooks/useUsers.ts
import { useMemo } from "react";
import { useQuery, QueryClient, QueryClientProvider } from "@tanstack/react-query";
import { UsersSchema, type User } from "../schemas/user";

// Discriminated union for explicit UI states
export type UsersState =
  | { status: "loading" }
  | { status: "error"; error: string }
  | { status: "success"; data: ReadonlyArray<User> };

// Fetch function with runtime validation and descriptive errors
async function fetchUsers(): Promise<ReadonlyArray<User>> {
  const res = await fetch("/api/users", { headers: { "accept": "application/json" } });
  if (!res.ok) {
    // Include status for observability; avoid leaking server internals
    throw new Error(`Request failed: ${res.status}`);
  }
  const json = await res.json();
  // Validate and coerce; throws if shape is wrong
  return UsersSchema.parse(json);
}

export function useUsers(): UsersState {
  const { data, error, status } = useQuery({
    queryKey: ["users"],
    queryFn: fetchUsers,
    staleTime: 60_000, // cache for 1 minute
    retry: 2,          // conservative retry policy
  });

  // Map TanStack Query status to a strict discriminated union for the UI
  return useMemo((): UsersState => {
    if (status === "pending") return { status: "loading" };
    if (status === "error") return { status: "error", error: (error as Error).message };
    // At this point data is defined and validated by Zod
    return { status: "success", data: data ?? [] };
  }, [status, error, data]);
}

// Optional: provide a QueryClient at the app root (copy-paste ready)
export const queryClient = new QueryClient();

// In your app root (e.g., src/main.tsx):
// import { createRoot } from "react-dom/client";
// import { QueryClientProvider } from "@tanstack/react-query";
// import { queryClient } from "./hooks/useUsers";
// import { App } from "./App";
// createRoot(document.getElementById("root")!).render(
//   <QueryClientProvider client={queryClient}>
//     <App />
//   </QueryClientProvider>
// );

3) Keep components presentational and accessible

// src/components/UsersList.tsx
import React from "react";
import { useUsers } from "../hooks/useUsers";

// Functional component focuses on rendering and accessibility
export function UsersList(): JSX.Element {
  const state = useUsers();

  if (state.status === "loading") {
    // Keep loading states lightweight and non-blocking
    return <p role="status" aria-live="polite">Loading users...</p>;
  }

  if (state.status === "error") {
    // Display a user-friendly message without revealing internals
    return <p role="alert">Could not load users. Please try again.</p>;
  }

  // Success path: minimal, semantic markup
  return (
    <ul aria-label="Users">
      {state.data.map(u => (
        <li key={u.id}>{u.name} ({u.email})</li>
      ))}
    </ul>
  );
}

4) Optional Azure back end: least-privilege, Managed Identity

This example shows an Azure Functions .NET 8 HTTP API that returns users. It authenticates to Azure Cosmos DB using DefaultAzureCredential and a system-assigned Managed Identity, avoiding connection strings. Assign only the necessary RBAC role.

// FunctionApp/Program.cs (.NET 8 isolated worker)
using Azure.Identity; // DefaultAzureCredential
using Microsoft.Azure.Functions.Worker;
using Microsoft.Azure.Functions.Worker.Http;
using Microsoft.Extensions.Hosting;
using Microsoft.Extensions.DependencyInjection;
using System.Net;
using Azure.Core;
using Azure.Cosmos; // Azure.Data.Cosmos alternative for .NET SDK

var host = new HostBuilder()
    .ConfigureFunctionsWorkerDefaults()
    .ConfigureServices(services =>
    {
        // Use managed identity via DefaultAzureCredential
        services.AddSingleton<TokenCredential>(_ => new DefaultAzureCredential());

        services.AddSingleton<CosmosClient>(sp =>
        {
            var credential = sp.GetRequiredService<TokenCredential>();
            // Endpoint from configuration (no keys). Use App Settings.
            var endpoint = Environment.GetEnvironmentVariable("COSMOS_ENDPOINT")!;
            return new CosmosClient(endpoint, credential);
        });
    })
    .Build();

await host.RunAsync();

// FunctionApp/GetUsers.cs
using Microsoft.Azure.Functions.Worker;
using Microsoft.Azure.Functions.Worker.Http;
using System.Net;
using Azure.Cosmos;
using System.Text.Json;

namespace FunctionApp;

public class GetUsers(CosmosClient cosmos)
{
    // HTTP-triggered function returning JSON users
    [Function("GetUsers")] 
    public async Task<HttpResponseData> Run(
        [HttpTrigger(AuthorizationLevel.Anonymous, "get", Route = "users")] HttpRequestData req)
    {
        var db = cosmos.GetDatabase("app");
        var container = db.GetContainer("users");

        var iterator = container.GetItemQueryIterator<UserDoc>("SELECT c.id, c.email, c.name FROM c");
        var results = new List<UserDoc>();
        while (iterator.HasMoreResults)
        {
            var page = await iterator.ReadNextAsync();
            results.AddRange(page);
        }

        var res = req.CreateResponse(HttpStatusCode.OK);
        await res.WriteStringAsync(JsonSerializer.Serialize(results));
        res.Headers.Add("Content-Type", "application/json");
        return res;
    }
}

public record UserDoc(string id, string email, string name);

Required RBAC (principle of least privilege): Assign the Function App's system-assigned identity the Cosmos DB Built-in Data Reader role scoped to the specific database or container. Avoid account-level permissions.

5) Minimal IaC for role assignment (Azure Bicep)

// main.bicep: create role assignment for Function App's managed identity
param cosmosAccountId string
param databaseRid string // scope appropriately (e.g., database or container resource ID)
param functionPrincipalId string // Function App's system-assigned identity principalId

resource roleDefinition 'Microsoft.Authorization/roleDefinitions@2022-04-01' existing = {
  scope: subscription()
  name: '00000000-0000-0000-0000-000000000001' // placeholder, replace with Cosmos DB Built-in Data Reader GUID
}

resource roleAssignment 'Microsoft.Authorization/roleAssignments@2022-04-01' = {
  name: guid(functionPrincipalId, databaseRid, roleDefinition.name)
  scope: resourceId('Microsoft.DocumentDB/databaseAccounts/sqlDatabases', cosmosAccountId, 'app')
  properties: {
    roleDefinitionId: roleDefinition.id
    principalId: functionPrincipalId
    principalType: 'ServicePrincipal'
  }
}

Note: Use azd to provision and configure environment variables like COSMOS_ENDPOINT. Never embed secrets or connection strings in code.

6) Wire up the React client to the Azure Function

// src/hooks/useUsers.ts (override the URL to your deployed Function App)
async function fetchUsers(): Promise<ReadonlyArray<User>> {
  const res = await fetch(import.meta.env.VITE_API_BASE + "/users", {
    headers: { accept: "application/json" },
    credentials: "include", // if using auth; otherwise omit
  });
  if (!res.ok) throw new Error(`Request failed: ${res.status}`);
  const json = await res.json();
  return UsersSchema.parse(json);
}

7) Testing hooks and components separately

// tests/useUsers.test.tsx
import { describe, it, expect } from "vitest";
import { QueryClient, QueryClientProvider } from "@tanstack/react-query";
import { renderHook, waitFor } from "@testing-library/react";
import { useUsers } from "../src/hooks/useUsers";

function wrapper({ children }: { children: React.ReactNode }) {
  const client = new QueryClient();
  return <QueryClientProvider client={client}>{children}</QueryClientProvider>;
}

describe("useUsers", () => {
  it("returns success after fetching", async () => {
    global.fetch = async () => new Response(JSON.stringify([]), { status: 200 });
    const { result } = renderHook(() => useUsers(), { wrapper });

    await waitFor(() => {
      expect(result.current.status).toBe("success");
    });
  });
});
// tests/UsersList.test.tsx
import { describe, it, expect } from "vitest";
import { render, screen } from "@testing-library/react";
import { QueryClient, QueryClientProvider } from "@tanstack/react-query";
import { UsersList } from "../src/components/UsersList";

function renderWithQuery(ui: React.ReactElement) {
  const client = new QueryClient();
  return render(<QueryClientProvider client={client}>{ui}</QueryClientProvider>);
}

describe("UsersList", () => {
  it("renders loading state", () => {
    renderWithQuery(<UsersList />);
    expect(screen.getByRole("status")).toHaveTextContent(/loading/i);
  });
});

Best Practices & Security

Hooks own side effects; components remain pure and predictable. Validate all external data with Zod and use strict TypeScript to catch issues at compile time. For Azure, prefer Managed Identity with DefaultAzureCredential and apply the smallest RBAC scope required. Keep API base URLs and configuration in environment variables managed by azd or your CI/CD system, not in source. For database reads with Entity Framework in .NET APIs, use AsNoTracking() to avoid unnecessary change tracking.

Summary

Hooks encapsulate reusable logic, I/O, and validation, while components render UI and stay testable. Strong typing with Zod and discriminated unions keeps state explicit and safe. Azure integration is secure with Managed Identity, least-privilege RBAC, and IaC via azd or Bicep.

Saturday, 24 January 2026

Avoiding the Pitfalls of Stopping Permission Inheritance: Performance Hotspots, Safer Patterns, and Azure-First Remediation

Stopping inherit permission (i.e., breaking permission inheritance) often seems like a quick fix for access control, but it introduces hidden operational and performance costs. This article explains why breaking inheritance in SharePoint and Azure RBAC leads to complexity, where performance issues occur, and how to remediate with .NET 8, TypeScript, and Azure-first patterns using Managed Identities and Infrastructure as Code.

The Problem

Breaking inheritance creates many one-off permission entries. Over time, this causes:

  • Permission sprawl: hard-to-audit, hard-to-revoke access scattered across items/resources.
  • Performance degradation: larger ACL evaluations, slower queries, and increased throttling risk.
  • Operational friction: brittle reviews, noisy exceptions, and confusing user experiences.

In SharePoint, many uniquely permissioned items slow list queries and complicate sharing. In Azure, assigning roles at leaf scopes (instead of using inherited assignments at management group or subscription levels) increases evaluation overhead and management burden.

Prerequisites

  • .NET 8 SDK
  • Node.js v20+
  • Azure CLI (az) and Azure Bicep
  • Contributor access to a test subscription (for deploying IaC) and Reader/Authorization permissions for audit scenarios

The Solution (Step-by-Step)

Step 1: What to avoid when stopping inheritance

  • SharePoint: Avoid per-item unique permissions for large lists; prefer groups at the site or library level, with exceptions gated by policy and approval.
  • Azure RBAC: Avoid many role assignments at the resource/resource-group level; grant least privilege at a higher scope when practical, and use groups instead of direct user assignments.

Step 2: .NET 8 Minimal API to detect RBAC hotspots (top-level statements, DI, Managed Identity)

// File: Program.cs (.NET 8, top-level statements, minimal API)
// Purpose: Audit Azure RBAC for non-inherited, leaf-level role assignments (hotspots).
// Auth: Uses DefaultAzureCredential (Managed Identity preferred in Azure).
// Note: This sample reads role assignments and surfaces "leaf" hotspots for review.

using Azure;
using Azure.Core;
using Azure.Identity;
using Azure.ResourceManager;
using Azure.ResourceManager.Authorization;
using Azure.ResourceManager.Authorization.Models;
using Microsoft.AspNetCore.Builder;
using Microsoft.Extensions.DependencyInjection;
using Microsoft.Extensions.Hosting;

var builder = WebApplication.CreateBuilder(args);

// Register Azure clients via DI
builder.Services.AddSingleton<TokenCredential>(_ => new DefaultAzureCredential());
// ArmClient is the entry point to Resource Manager APIs
builder.Services.AddSingleton<ArmClient>(sp => new ArmClient(sp.GetRequiredService<TokenCredential>()));

var app = builder.Build();

// GET /rbac/hotspots?subscriptionId=<subId>
// Heuristic: Identify role assignments at deeper scopes (resource, resource group) that
// could be consolidated at higher scopes to reduce sprawl and evaluation overhead.
app.MapGet("/rbac/hotspots", async (string subscriptionId, ArmClient arm) =>
{
    // Acquire subscription resource
    var sub = arm.GetSubscriptionResource(new ResourceIdentifier($"/subscriptions/{subscriptionId}"));

    // List role assignments across the subscription
    var assignments = new List<RoleAssignmentData>();
    await foreach (var ra in sub.GetRoleAssignmentsAsync())
    {
        assignments.Add(ra.Data);
    }

    // Group by scope depth (subscription=1, resourceGroup=2, resource=3+)
    // Deeper scopes are more likely to be "breaks" from inheritance-like patterns.
    var hotspots = assignments
        .GroupBy(a => ScopeDepth(a.Scope))
        .OrderByDescending(g => g.Key)
        .Select(g => new
        {
            depth = g.Key,
            count = g.Count(),
            sampleScopes = g.Select(x => x.Scope).Distinct().Take(5).ToArray()
        });

    return Results.Ok(new
    {
        analyzed = assignments.Count,
        hotspots
    });

    // Simple helper to score scope depth by path segments.
    static int ScopeDepth(string scope)
    {
        // Example: /subscriptions/{id} => depth ~ 1
        // /subscriptions/{id}/resourceGroups/{rg} => depth ~ 2
        // /subscriptions/{id}/resourceGroups/{rg}/providers/... => depth >= 3
        return scope.Split('/', StringSplitOptions.RemoveEmptyEntries).Length / 2;
    }
})
.WithName("GetRbacHotspots")
.Produces<object>(200);

app.Run();

Why this helps: Large counts of deep-scope assignments often indicate broken inheritance patterns (per-resource grants). Consolidating to group-based roles at higher scopes can reduce policy evaluation work and administrative overhead.

Step 3: TypeScript Azure Function to list deep-scope role assignments with retry (Managed Identity)

// File: index.ts (Azure Functions v4, Node 20)
// Purpose: Enumerate role assignments and flag deep-scope patterns.
// Auth: ManagedIdentityCredential (no client secrets). Strict typing. Basic retry for 429.
//
// Ensure you enable a system-assigned identity on the Function App and grant it Reader on the subscription.
// package.json should include: @azure/identity, @azure/arm-authorization, zod

import { AzureFunction, Context } from "@azure/functions";
import { ManagedIdentityCredential } from "@azure/identity";
import { AuthorizationManagementClient, RoleAssignment } from "@azure/arm-authorization";
import { z } from "zod";

// Validate required environment variable via Zod (strict typing)
const EnvSchema = z.object({
  SUBSCRIPTION_ID: z.string().min(1)
});
const env = EnvSchema.parse(process.env);

const httpTrigger: AzureFunction = async function (context: Context): Promise<void> {
  // Create credential using Managed Identity (no secrets in code or env)
  const credential = new ManagedIdentityCredential();
  const authClient = new AuthorizationManagementClient(credential, env.SUBSCRIPTION_ID);

  // Simple retry wrapper for throttle-prone calls (e.g., large tenants)
  async function withRetry<T>(fn: () => Promise<T>, attempts = 5, delayMs = 1000): Promise<T> {
    let lastErr: unknown;
    for (let i = 0; i < attempts; i++) {
      try {
        return await fn();
      } catch (err: unknown) {
        const anyErr = err as { statusCode?: number };
        if (anyErr?.statusCode === 429 || anyErr?.statusCode === 503) {
          await new Promise((r) => setTimeout(r, delayMs * (i + 1))); // Exponential-ish backoff
          lastErr = err;
          continue;
        }
        throw err;
      }
    }
    throw lastErr;
  }

  // List role assignments at subscription scope
  const scope = `/subscriptions/${env.SUBSCRIPTION_ID}`;
  const assignments: RoleAssignment[] = [];

  // Use retry around list calls; the SDK returns an async iterator
  const pager = authClient.roleAssignments.listForSubscription();
  for await (const item of pager) {
    assignments.push(item);
  }

  // Score depth by scope path complexity
  const scoreDepth = (s: string): number => s.split("/").filter(Boolean).length / 2;
  const hotspots = assignments
    .map(a => ({ id: a.id!, scope: a.scope!, depth: scoreDepth(a.scope!) }))
    .filter(x => x.depth >= 3); // resource-level assignments

  context.res = {
    status: 200,
    headers: { "content-type": "application/json" },
    body: {
      analyzed: assignments.length,
      resourceLevelAssignments: hotspots.length,
      hotspotSamples: hotspots.slice(0, 10)
    }
  };
};

export default httpTrigger;

Why this helps: A quick serverless audit allows teams to discover where inheritance-like patterns are being bypassed in Azure RBAC, which is a frequent source of performance and governance friction.

Step 4: Azure Bicep to deploy a Function App with Managed Identity and least privilege

// File: main.bicep
// Purpose: Deploy a Function App with system-assigned managed identity,
// and assign Reader at the resource group scope (principle of least privilege).
// Includes a deterministic GUID for role assignment name.
//
// Note: Reader role definition ID is acdd72a7-3385-48ef-bd42-f606fba81ae7
// (Allows viewing resources, not making changes.)

@description('Location for resources')
param location string = resourceGroup().location

@description('Function App name')
param functionAppName string

@description('Storage account name (must be globally unique)')
param storageAccountName string

var roleDefinitionIdReader = '/providers/Microsoft.Authorization/roleDefinitions/acdd72a7-3385-48ef-bd42-f606fba81ae7'

// Storage (for Functions)
resource stg 'Microsoft.Storage/storageAccounts@2023-01-01' = {
  name: storageAccountName
  location: location
  sku: {
    name: 'Standard_LRS'
  }
  kind: 'StorageV2'
  properties: {
    allowBlobPublicAccess: false
    minimumTlsVersion: 'TLS1_2'
    supportsHttpsTrafficOnly: true
  }
}

// Hosting plan (Consumption)
resource plan 'Microsoft.Web/serverfarms@2022-09-01' = {
  name: '${functionAppName}-plan'
  location: location
  sku: {
    name: 'Y1'
    tier: 'Dynamic'
  }
}

// Function App with system-assigned identity
resource func 'Microsoft.Web/sites@2022-09-01' = {
  name: functionAppName
  location: location
  kind: 'functionapp'
  identity: {
    type: 'SystemAssigned'
  }
  properties: {
    serverFarmId: plan.id
    siteConfig: {
      appSettings: [
        { name: 'FUNCTIONS_WORKER_RUNTIME', value: 'node' }
        { name: 'WEBSITE_RUN_FROM_PACKAGE', value: '1' }
      ]
    }
  }
}

// Assign Reader at the resource group scope to the Function's identity
// Use a stable, deterministic GUID based on scope + principal to avoid duplicates.
resource readerAssign 'Microsoft.Authorization/roleAssignments@2022-04-01' = {
  name: guid(resourceGroup().id, func.identity.principalId, roleDefinitionIdReader)
  scope: resourceGroup()
  properties: {
    roleDefinitionId: subscriptionResourceId('Microsoft.Authorization/roleDefinitions', 'acdd72a7-3385-48ef-bd42-f606fba81ae7')
    principalId: func.identity.principalId
    principalType: 'ServicePrincipal'
  }
}

Why this helps: You deploy secure defaults and enforce least privilege by design. The deterministic GUID prevents accidental duplicate role assignments. Reader is explicitly chosen to avoid write permissions while enabling inventory and audits.

Where performance issues occur

  • SharePoint: Many items with unique permissions increase ACL checks and can slow list queries, indexing, and certain sharing operations. Batch operations on items with unique permissions are more likely to hit throttling.
  • Azure RBAC: Thousands of per-resource role assignments increase evaluation work during authorization and complicate policy and compliance scans. It also prolongs investigations during incidents.
  • Auditing and reviews: Per-user, per-resource assignments inflate review surfaces and make access recertification slow and error-prone.

Best Practices & Security

  • Use Managed Identity or Azure AD Workload Identity. Avoid client secrets for server-side workloads. Do not store secrets in environment variables. If you must handle secrets, use Azure Key Vault with RBAC and Managed Identity.
  • Prefer group-based assignments over direct user assignments. This simplifies reviews and minimizes churn.
  • Favor higher-scope role assignments with least privilege. Start at management group or subscription only when justified and narrow to Reader or custom roles that fit the minimal required actions.
  • When exceptions are necessary, document them. Add expiration and owners to each exception.
  • For SharePoint, grant permissions at the site or library level using groups. Reserve item-level breaks for rare, time-bound cases.
  • Monitor continuously. Integrate with Azure Monitor and Azure Policy to detect excessive deep-scope assignments. Create alerts on abnormal growth of role assignments or access anomalies.
  • Implement retry and backoff for API calls that can throttle (429/503), both in audits and operational tooling.
  • Standardize terminology. Use "inheritance" consistently to avoid confusion in documentation and automation.

Recommendation: If you need read-only inventory across a subscription, assign the Reader role (roleDefinitionId acdd72a7-3385-48ef-bd42-f606fba81ae7) to a Managed Identity and call Azure SDKs or ARM REST with exponential backoff.

Summary

  • Breaking inheritance increases complexity and can degrade performance; reserve it for rare, time-bound exceptions.
  • Automate audits with Managed Identity using .NET 8 and TypeScript to find deep-scope hotspots and consolidate access.
  • Ship secure-by-default with Bicep: least privilege (Reader), deterministic role assignment IDs, and continuous monitoring via Azure services.

Friday, 23 January 2026

SharePoint Indexing: A Step-by-Step Guide for Faster Search and Precise Results

SharePoint indexing often fails silently, causing slow search and missing results. This guide shows exactly how to configure SharePoint indexing end-to-end—site settings, list-level indexing, search schema, and safe automation—so your users get fast, accurate results. Primary keyword: SharePoint indexing.

The Problem

Content updates are not discoverable, queries are slow, and filters return inconsistent items because lists aren’t indexed, sites aren’t reindexed, or managed properties are not mapped. You need a repeatable approach to fix indexing across sites and automate reindexing safely.

Prerequisites

  • Microsoft 365 tenant with SharePoint Online
  • PowerShell 7+ and PnP.PowerShell module (Install-Module PnP.PowerShell -Scope CurrentUser)
  • Role-based access (least privilege):
  • Site Owner or above for list/site indexing and reindex
  • Search Schema changes typically require SharePoint Administrator
  • Automation with app-only: Entra ID App with Sites.Selected + site-scoped permissions
  • Optional for automation: Azure Automation or Azure Functions with a System-Assigned Managed Identity

The Solution (Step-by-Step)

1) Confirm site-level indexing is allowed

Site Owners can check if content is visible to search crawlers.

  • Go to Site Settings → Search → Search and offline availability
  • Set Allow this site to appear in search results to Yes

Pro-Tip: If you changed this from No to Yes, trigger a site reindex to speed up propagation.

2) Index columns at the list/library level

Index columns used in filters, sorts, and query rules to avoid throttling and improve query performance.

  • Open your list/library → Settings → Indexed columns → Create a new index
  • Index only frequently queried columns (e.g., Status, Department, Created)

Pro-Tip: Avoid indexing lookup or multi-line rich text columns unless absolutely necessary; they can increase index size and crawl churn.

3) Reindex a list or library (targeted)

Use targeted reindex after column or schema changes on a specific list.

  • List Settings → Advanced settings → Reindex List

This flags the list for the next crawl so updated properties and new mappings are applied faster.

4) Reindex a site (broad fix)

Use this when you changed site-level search settings, content types, or many lists.

  • Site Settings → Search → Search and offline availability → Reindex site

Pro-Tip: Reindexing a site is heavier than a list reindex. Prefer list reindex when possible to reduce crawl load.

5) Map crawled properties to managed properties (search schema)

To make custom columns queryable/filterable/sortable, map crawled properties to managed properties with the right search attributes (Searchable, Queryable, Retrievable, Refinable, Sortable).

  • SharePoint Admin Center → More features → Search → Open
  • Manage Search Schema → Managed Properties
  • Create or edit a managed property (e.g., RefinableStringXX)
  • Map the relevant crawled property (e.g., ows_Status)

Pro-Tip: Use reserved RefinableStringXX or RefinableDateXX for facets and filters to avoid schema conflicts.

6) Automate reindex with PnP.PowerShell (interactive)

# Install PnP.PowerShell if needed
# Install-Module PnP.PowerShell -Scope CurrentUser

# 1) Interactive login for an admin or site owner
Connect-PnPOnline -Url "https://contoso.sharepoint.com/sites/ProjectA" -Interactive

# 2) Reindex a single list by title
# This sets the reindex flag so the next crawl refreshes properties
$web = Get-PnPWeb
$list = Get-PnPList -Identity "Documents"
Set-PnPList -Identity $list -NoCrawl:$false   # Ensure list is crawlable
Request-PnPReIndexList -Identity $list        # Mark list for reindex

# 3) Reindex the entire site (use sparingly)
Request-PnPReIndexWeb

# 4) Verify a column is indexed
# Returns indexed column definitions for the list
(Get-PnPField -List $list) | Where-Object { $_.Indexed -eq $true } | Select-Object InternalName, Title

Comments:

  • Request-PnPReIndexList and Request-PnPReIndexWeb mark content for recrawl.
  • Set-PnPList -NoCrawl:$false ensures the list is included in search.
  • Use least privilege: Site Owner is sufficient for list-level operations.

7) Secure automation with Managed Identity or app-only (no secrets)

For production jobs, avoid interactive auth and stored passwords. Use a Managed Identity (Azure Automation/Functions) or an Entra ID app with Sites.Selected and scoped permissions to specific sites.

# Option A: Managed Identity (runs inside Azure with System-Assigned MI)
# Prereqs:
# - Assign the Managed Identity Sites.Selected permissions and grant site-level rights
# - Use Grant-PnPAzureADAppSitePermission to scope access to the target site

# Connect using Managed Identity
Connect-PnPOnline -Url "https://contoso.sharepoint.com/sites/ProjectA" -ManagedIdentity

# Reindex specific list
Request-PnPReIndexList -Identity "Documents"
# Option B: Entra ID App with certificate (app-only, least privilege)
# Prereqs:
# - App registration with Microsoft Graph Sites.Selected (Application) permission
# - Admin consent granted
# - Grant site-level permission:
#   Grant-PnPAzureADAppSitePermission -AppId <CLIENT_ID> -DisplayName "Indexer" -Site "https://contoso.sharepoint.com/sites/ProjectA" -Permissions Read
# - Upload certificate and use its thumbprint

$tenant = "contoso.onmicrosoft.com"
$siteUrl = "https://contoso.sharepoint.com/sites/ProjectA"
$clientId = "00000000-0000-0000-0000-000000000000"
$thumb = "THUMBPRINT_HERE"

Connect-PnPOnline -Url $siteUrl -ClientId $clientId -Tenant $tenant -Thumbprint $thumb

# Safely trigger reindex
Request-PnPReIndexList -Identity "Documents"

Comments:

  • Sites.Selected lets you grant per-site permissions, enforcing least privilege.
  • Avoid legacy ACS app-only. Prefer Entra ID + Sites.Selected or Managed Identity.
  • No secrets in code; use certificates or Managed Identity.

8) Validate search consistency

  • Use Microsoft Search to query values you expect to find (e.g., Status:Active)
  • Validate refiners show the expected values (requires RefinableStringXX mapping)
  • Check list view performance with indexed columns applied as filters

Best Practices & Security

  • Principle of Least Privilege:
  • Day-to-day indexing: Site Owner
  • Search Schema: SharePoint Administrator (only when needed)
  • Automation: Managed Identity or app with Sites.Selected scoped to the target sites
  • Authentication:
  • Prefer Managed Identity in Azure Automation/Functions
  • Else use certificate-based app-only; avoid passwords and client secrets
  • Search Schema Hygiene:
  • Reuse RefinableStringXX/DateXX for facets and filters
  • Document managed property usage to prevent collisions across teams
  • Performance:
  • Index only columns that improve key queries
  • Prefer list reindex over site reindex to reduce crawl load

Pro-Tip: Create a small pilot list to test new mappings and reindex timings before applying to large libraries.

Troubleshooting

  • My column values don’t appear in search:
  • Confirm list is crawlable (NoCrawl = false)
  • Ensure a managed property is mapped and set to Queryable/Retrievable
  • Trigger Request-PnPReIndexList and wait for the next crawl cycle
  • Refiners don’t show my custom metadata:
  • Use a RefinableStringXX or RefinableDateXX managed property
  • Map the correct crawled property (often ows_ColumnInternalName)
  • App-only connection fails with 401/403:
  • Verify Sites.Selected consent and site-level grant (Grant-PnPAzureADAppSitePermission)
  • Confirm certificate thumbprint and validity
  • Reindex seems to do nothing:
  • Allow time for the next crawl; reindex sets a flag, it doesn’t force immediate recrawl
  • Check Service Health and Message Center for crawl incidents

Summary

  • Index the right columns and map to managed properties to make content queryable, refinable, and fast.
  • Use targeted list reindex first; reserve site reindex for broad changes.
  • Automate safely with Managed Identity or Sites.Selected app-only to enforce least privilege and avoid secrets.

References

Sunday, 18 January 2026

Integrate Power Apps with AI using Azure Functions (.NET 8) and Azure OpenAI with Managed Identity

Integrate Power Apps with AI by fronting Azure OpenAI behind a secure .NET 8 isolated Azure Function, authenticated via Entra ID and deployed with Azure Bicep. Why this matters: you keep your Azure OpenAI keyless via Managed Identity, enforce RBAC, and provide a stable HTTPS endpoint for Power Apps using a custom connector.

The Problem

Developers need to call AI securely from Power Apps without exposing keys, while meeting enterprise requirements for RBAC, observability, and least-privilege. Manual wiring through the portal and ad hoc security checks lead to drift and risk.

Prerequisites

  • Azure CLI 2.60+
  • .NET 8 SDK
  • Azure Developer CLI (optional) or Bicep CLI
  • Owner or User Access Administrator on the target subscription
  • Power Apps environment access for creating a custom connector

The Solution (Step-by-Step)

1) Deploy Azure resources with Bicep (Managed Identity, Function App, Azure OpenAI, RBAC, Easy Auth)

This Bicep template creates a Function App with system-assigned managed identity, Azure OpenAI with a model deployment, App Service Authentication (Easy Auth) enforced, and assigns the Cognitive Services OpenAI User role to the identity.

// main.bicep
targetScope = 'resourceGroup'

@description('Name prefix for resources')
param namePrefix string

@description('Location for resources')
param location string = resourceGroup().location

@description('Azure OpenAI model name, e.g., gpt-4o-mini')
param aoaiModelName string = 'gpt-4o-mini'

@description('Azure OpenAI model version for your region (see Azure docs for supported versions).')
param aoaiModelVersion string = '2024-07-18' // Tip: Validate the correct version via az cognitiveservices account list-models

var funcName = '${namePrefix}-func'
var aoaiName = '${namePrefix}-aoai'
var hostingPlanName = '${namePrefix}-plan'
var appInsightsName = '${namePrefix}-ai'
var storageName = toLower(replace('${namePrefix}st${uniqueString(resourceGroup().id)}', '-', ''))

resource storage 'Microsoft.Storage/storageAccounts@2023-05-01' = {
  name: storageName
  location: location
  sku: {
    name: 'Standard_LRS'
  }
  kind: 'StorageV2'
}

resource appInsights 'Microsoft.Insights/components@2020-02-02' = {
  name: appInsightsName
  location: location
  kind: 'web'
  properties: {
    Application_Type: 'web'
  }
}

resource plan 'Microsoft.Web/serverfarms@2023-12-01' = {
  name: hostingPlanName
  location: location
  sku: {
    name: 'Y1' // Consumption for Functions
    tier: 'Dynamic'
  }
}

resource func 'Microsoft.Web/sites@2023-12-01' = {
  name: funcName
  location: location
  kind: 'functionapp'
  identity: {
    type: 'SystemAssigned'
  }
  properties: {
    serverFarmId: plan.id
    siteConfig: {
      appSettings: [
        { name: 'AzureWebJobsStorage', value: storage.listKeys().keys[0].value }
        { name: 'APPLICATIONINSIGHTS_CONNECTION_STRING', value: appInsights.properties.ConnectionString }
        { name: 'FUNCTIONS_EXTENSION_VERSION', value: '~4' }
        { name: 'FUNCTIONS_WORKER_RUNTIME', value: 'dotnet-isolated' }
        // Enable Easy Auth enforcement via config below (authSettingsV2)
      ]
      http20Enabled: true
      minimumTlsVersion: '1.2'
      ftpsState: 'Disabled'
      cors: {
        allowedOrigins: [
          // Add your Power Apps domain if needed; CORS is typically managed by the custom connector
        ]
      }
    }
    httpsOnly: true
  }
}

// Enable App Service Authentication (Easy Auth) with Entra ID and enforce authentication.
resource auth 'Microsoft.Web/sites/config@2023-12-01' = {
  name: '${func.name}/authsettingsV2'
  properties: {
    globalValidation: {
      requireAuthentication: true
      unauthenticatedClientAction: 'Return401'
    }
    identityProviders: {
      azureActiveDirectory: {
        enabled: true
        registration: {
          // When omitted, system can use an implicit provider. For enterprise, specify a registered app:
          // Provide clientId and clientSecretSettingName if using a dedicated app registration.
        }
        validation: {
          // Audience (App ID URI or application ID). Power Apps custom connector should request this audience.
          // Replace with your API application ID URI when using a dedicated AAD app registration.
          allowedAudiences: [
            'api://<your-app-id>'
          ]
        }
        login: {
          disableWWWAuthenticate: false
        }
      }
    }
    platform: {
      enabled: true
      runtimeVersion: '~1'
    }
    login: {
      tokenStore: {
        enabled: true
      }
    }
  }
}

// Azure OpenAI account
resource aoai 'Microsoft.CognitiveServices/accounts@2023-05-01' = {
  name: aoaiName
  location: location
  kind: 'OpenAI'
  sku: {
    name: 'S0'
  }
  properties: {
    customSubDomainName: toLower(aoaiName)
    publicNetworkAccess: 'Enabled'
  }
}

// Azure OpenAI deployment - ensure model/version are valid for your region
resource aoaiDeployment 'Microsoft.CognitiveServices/accounts/deployments@2023-05-01' = {
  name: 'chat-${aoaiModelName}'
  parent: aoai
  properties: {
    model: {
      format: 'OpenAI'
      name: aoaiModelName
      version: aoaiModelVersion
    }
    capacities: [
      {
        capacity: 1
        capacityType: 'Standard'
      }
    ]
  }
}

// Assign RBAC: Cognitive Services OpenAI User role to the Function's managed identity
// Built-in role: Cognitive Services OpenAI User (Role ID: 5e0bd9bd-7ac1-4c9e-8289-1b01f135d4a8)
resource roleAssignment 'Microsoft.Authorization/roleAssignments@2022-04-01' = {
  name: guid(aoai.id, func.identity.principalId, 'CognitiveServicesOpenAIUserRole')
  scope: aoai
  properties: {
    roleDefinitionId: subscriptionResourceId('Microsoft.Authorization/roleDefinitions', '5e0bd9bd-7ac1-4c9e-8289-1b01f135d4a8')
    principalId: func.identity.principalId
    principalType: 'ServicePrincipal'
  }
}

output functionAppName string = func.name
output openAiEndpoint string = 'https://${aoai.name}.openai.azure.com/'
output openAiDeployment string = aoaiDeployment.name

Tip: To find the correct model version for your region, run: az cognitiveservices account list-models --name <aoaiName> --resource-group <rg> --output table.

2) Implement a secure .NET 8 isolated Azure Function with DI and Managed Identity

The function validates Entra ID via Easy Auth headers, performs input validation, calls Azure OpenAI using DefaultAzureCredential, and returns a minimal response. AuthorizationLevel is Anonymous because App Service Authentication is enforcing auth at the edge; the code still verifies identity defensively.

// Program.cs
using System.Text.Json;
using Azure;
using Azure.AI.OpenAI;
using Azure.Identity;
using Microsoft.Azure.Functions.Worker;
using Microsoft.Azure.Functions.Worker.Middleware;
using Microsoft.Azure.Functions.Worker.Extensions.OpenApi.Extensions;
using Microsoft.Extensions.Configuration;
using Microsoft.Extensions.DependencyInjection;
using Microsoft.Extensions.Hosting;
using Microsoft.Extensions.Logging;

var host = new HostBuilder()
    .ConfigureFunctionsWebApplication(builder =>
    {
        // Add global middleware if needed (e.g., correlation, exception handling)
        builder.UseMiddleware<GlobalExceptionMiddleware>();
    })
    .ConfigureServices((ctx, services) =>
    {
        var config = ctx.Configuration;

        // Bind custom options
        services.Configure<OpenAIOptions>(config.GetSection(OpenAIOptions.SectionName));

        // Register OpenAI client using Managed Identity (no keys)
        services.AddSingleton((sp) =>
        {
            var options = sp.GetRequiredService<Microsoft.Extensions.Options.IOptions<OpenAIOptions>>().Value;
            // Use DefaultAzureCredential to leverage Managed Identity in Azure
            var credential = new DefaultAzureCredential();
            return new OpenAIClient(new Uri(options.Endpoint), credential);
        });

        services.AddSingleton<IPromptService, PromptService>();
        services.AddApplicationInsightsTelemetryWorkerService();
    })
    .ConfigureAppConfiguration(config =>
    {
        config.AddEnvironmentVariables();
    })
    .ConfigureLogging(logging =>
    {
        logging.AddConsole();
    })
    .Build();

await host.RunAsync();

file sealed class GlobalExceptionMiddleware : IFunctionsWorkerMiddleware
{
    public async Task Invoke(FunctionContext context, FunctionExecutionDelegate next)
    {
        try
        {
            await next(context);
        }
        catch (Exception ex)
        {
            var logger = context.GetLogger<GlobalExceptionMiddleware>();
            logger.LogError(ex, "Unhandled exception");

            // Avoid leaking internals; return generic error with correlation ID
            var invocationId = context.InvocationId;
            var http = await context.GetHttpRequestDataAsync();
            if (http is not null)
            {
                var response = http.CreateResponse(System.Net.HttpStatusCode.InternalServerError);
                await response.WriteStringAsync($"Request failed. CorrelationId: {invocationId}");
                context.GetInvocationResult().Value = response;
            }
        }
    }
}

file sealed class OpenAIOptions
{
    public const string SectionName = "OpenAI";
    public string Endpoint { get; init; } = string.Empty; // e.g., https://<aoaiName>.openai.azure.com/
    public string Deployment { get; init; } = string.Empty; // e.g., chat-gpt-4o-mini
}

// Service abstraction to keep function logic clean
file interface IPromptService
{
    Task<string> CreateChatResponseAsync(string userInput, CancellationToken ct);
}

file sealed class PromptService(OpenAIClient client, Microsoft.Extensions.Options.IOptions<OpenAIOptions> opts, ILogger<PromptService> logger) : IPromptService
{
    private readonly OpenAIClient _client = client;
    private readonly OpenAIOptions _opts = opts.Value;
    private readonly ILogger<PromptService> _logger = logger;

    public async Task<string> CreateChatResponseAsync(string userInput, CancellationToken ct)
    {
        // Basic input trimming and minimal length check
        var prompt = (userInput ?? string.Empty).Trim();
        if (prompt.Length < 3)
        {
            return "Input too short.";
        }

        // Create chat completion with minimal system prompt
        var req = new ChatCompletionsOptions()
        {
            DeploymentName = _opts.Deployment,
            Temperature = 0.2f,
            MaxTokens = 256
        };
        req.Messages.Add(new ChatRequestSystemMessage("You are a concise assistant.")); // Guardrail
        req.Messages.Add(new ChatRequestUserMessage(prompt));

        _logger.LogInformation("Calling Azure OpenAI deployment {Deployment}", _opts.Deployment);

        var resp = await _client.GetChatCompletionsAsync(req, ct);
        var msg = resp.Value.Choices.Count > 0 ? resp.Value.Choices[0].Message.Content[0].Text : "No response.";
        return msg ?? string.Empty;
    }
}
// HttpFunction.cs
using System.Net;
using System.Security.Claims;
using System.Text;
using System.Text.Json;
using Microsoft.Azure.Functions.Worker;
using Microsoft.Azure.Functions.Worker.Http;
using Microsoft.Extensions.Logging;
using Microsoft.Extensions.Options;

namespace Api;

public sealed class ChatRequest
{
    // Strongly typed DTO for request validation
    public string Prompt { get; init; } = string.Empty;
}

public sealed class ChatFunction(IPromptService promptService, ILogger<ChatFunction> logger) 
{
    // Simple helper to read Easy Auth principal header safely
    private static ClaimsPrincipal? TryGetPrincipal(HttpRequestData req)
    {
        // Easy Auth sets x-ms-client-principal as Base64 JSON
        if (!req.Headers.TryGetValues("x-ms-client-principal", out var values)) return null;
        var b64 = values.FirstOrDefault();
        if (string.IsNullOrWhiteSpace(b64)) return null;

        try
        {
            var json = Encoding.UTF8.GetString(Convert.FromBase64String(b64));
            using var doc = JsonDocument.Parse(json);
            var claims = new List<Claim>();
            if (doc.RootElement.TryGetProperty("claims", out var arr))
            {
                foreach (var c in arr.EnumerateArray())
                {
                    var typ = c.GetProperty("typ").GetString() ?? string.Empty;
                    var val = c.GetProperty("val").GetString() ?? string.Empty;
                    claims.Add(new Claim(typ, val));
                }
            }
            return new ClaimsPrincipal(new ClaimsIdentity(claims, "EasyAuth"));
        }
        catch
        {
            return null;
        }
    }

    [Function("chat")]
    public async Task<HttpResponseData> Run(
        [HttpTrigger(AuthorizationLevel.Anonymous, "post", Route = "chat")] HttpRequestData req,
        FunctionContext context)
    {
        var principal = TryGetPrincipal(req);
        if (principal is null)
        {
            // Defense in depth: App Service auth should already block, but we double-check.
            var unauth = req.CreateResponse(HttpStatusCode.Unauthorized);
            await unauth.WriteStringAsync("Authentication required.");
            return unauth;
        }

        // Basic content-type check and bound read limit
        if (!req.Headers.TryGetValues("Content-Type", out var ct) || !ct.First().Contains("application/json", StringComparison.OrdinalIgnoreCase))
        {
            var bad = req.CreateResponse(HttpStatusCode.BadRequest);
            await bad.WriteStringAsync("Content-Type must be application/json.");
            return bad;
        }

        // Safe body read
        using var reader = new StreamReader(req.Body);
        var body = await reader.ReadToEndAsync();
        ChatRequest? input;
        try
        {
            input = JsonSerializer.Deserialize<ChatRequest>(body, new JsonSerializerOptions { PropertyNameCaseInsensitive = true });
        }
        catch
        {
            var bad = req.CreateResponse(HttpStatusCode.BadRequest);
            await bad.WriteStringAsync("Invalid JSON.");
            return bad;
        }

        if (input is null || string.IsNullOrWhiteSpace(input.Prompt))
        {
            var bad = req.CreateResponse(HttpStatusCode.BadRequest);
            await bad.WriteStringAsync("Prompt is required.");
            return bad;
        }

        var loggerScope = new Dictionary<string, object> { ["UserObjectId"] = principal.FindFirst("http://schemas.microsoft.com/identity/claims/objectidentifier")?.Value ?? "unknown" };
        using (logger.BeginScope(loggerScope))
        {
            try
            {
                var result = await promptService.CreateChatResponseAsync(input.Prompt, context.CancellationToken);
                var ok = req.CreateResponse(HttpStatusCode.OK);
                await ok.WriteAsJsonAsync(new { response = result });
                return ok;
            }
            catch (Exception ex)
            {
                logger.LogError(ex, "Error generating chat response");
                var err = req.CreateResponse(HttpStatusCode.BadGateway);
                await err.WriteStringAsync("AI request failed. Try again later.");
                return err;
            }
        }
    }
}
// appsettings.json (for local debug)
{
  "OpenAI": {
    "Endpoint": "https://<your-aoai-name>.openai.azure.com/",
    "Deployment": "chat-gpt-4o-mini"
  }
}

Tip: When running locally, ensure you are logged in with Azure CLI (az login) so DefaultAzureCredential can acquire a token. Assign yourself Cognitive Services OpenAI User on the AOAI resource for testing.

3) Enforce Azure AD (Easy Auth) - configuration example

The Bicep above enables Easy Auth and returns 401 for unauthenticated calls. If you maintain settings as JSON, you can also use an ARM style configuration:

// Example snippet (appsettings for site config via ARM/Bicep-style) ensures:
// - requireAuthentication: true
// - allowedAudiences must match the audience used by your client (Power Apps custom connector)
{
  "properties": {
    "authSettingsV2": {
      "platform": { "enabled": true },
      "globalValidation": {
        "requireAuthentication": true,
        "unauthenticatedClientAction": "Return401"
      },
      "identityProviders": {
        "azureActiveDirectory": {
          "enabled": true,
          "validation": {
            "allowedAudiences": [ "api://<your-app-id>" ]
          }
        }
      }
    }
  }
}

Tip: For most enterprise setups, register a dedicated App Registration for the Function API and reference its Application ID URI in allowedAudiences. Keep consent and scopes explicit.

4) Create a Power Apps Custom Connector

  1. In Power Apps, open Solutions and create a new Custom Connector.
  2. Set the Host to your Function URL (e.g., https://<func-name>.azurewebsites.net).
  3. Define the POST /chat operation that accepts a JSON body { "prompt": "..." } and returns { "response": "..." }.
  4. Security: Choose OAuth 2.0 (Azure Active Directory). Set Audience to match allowedAudiences (api://<your-app-id>). Supply Tenant ID and Client ID as provided by your admin.
  5. Test the connector. You should receive HTTP 200 with the AI response. Unauthorized calls return 401.

5) Add client-side validation in Power Apps (Power Fx)

Before invoking the connector, validate inputs in the app to fail fast and reduce server load.

// Example Power Fx for a button's OnSelect
If(
    IsBlank(Trim(txtPrompt.Text)) || Len(Trim(txtPrompt.Text)) < 3,
    Notify("Please enter at least 3 characters.", NotificationType.Error),
    Set(
        aiResult,
        YourConnector.chat(
            {
                prompt: Trim(txtPrompt.Text)
            }
        ).response
    )
);

// Display result in a Label: Text = aiResult

Tip: Add input length limits and debouncing for multi-keystroke actions. For read-only UI against data, prefer server-side pagination and AsNoTracking() on the API layer where EF Core is involved.

Best Practices & Security

Identity and RBAC

  • Use Managed Identity for the Function App; never store secrets. The sample uses DefaultAzureCredential.
  • Assign the Cognitive Services OpenAI User role to the Function’s identity on the Azure OpenAI resource. This allows API calls without exposing keys.
  • Enforce Azure AD with Easy Auth at the platform level and verify identity in code for defense in depth.

Least Privilege and Network

  • Restrict access with the minimum RBAC scope necessary. Avoid granting Contributor on the subscription.
  • Enable HTTPS only and TLS 1.2+. Disable FTP/FTPS where possible.
  • Consider Private Endpoints for Azure OpenAI and Functions behind an Application Gateway/WAF for enterprise networks.

Validation, Error Handling, and Reliability

  • Validate input on both client (Power Apps) and server (Function). The code enforces content-type and schema shape.
  • Handle exceptions using middleware and return generic messages with correlation IDs to avoid leaking internals.
  • Implement circuit breakers and retries for downstream calls with transient fault policies if you add HTTP dependencies.

Observability

  • Use Application Insights. Log prompt sizes and latency, not raw sensitive content.
  • Sample KQL to investigate errors:
// Requests by operation
requests
| where url endswith "/chat"
| summarize count() by resultCode

// Exceptions with correlation
exceptions
| where operation_Name == "chat"
| project timestamp, type, problemId, outerMessage, operation_Id

// Latency p95
requests
| where url endswith "/chat"
| summarize p95(duration) by bin(timestamp, 1h)

Cost and Abuse Controls

  • Apply rate limiting at the API Management layer if exposing broadly.
  • Set MaxTokens and Temperature conservatively; monitor usage with budgets and alerts.
  • Implement content filters as needed using Azure AI Content Safety.

Summary

  • Power Apps integrates cleanly with AI by calling a secure .NET 8 Azure Function that uses Managed Identity to access Azure OpenAI.
  • Security is enforced with Azure AD (Easy Auth) and RBAC, specifically the Cognitive Services OpenAI User role.
  • IaC with Bicep, defensive coding, and Application Insights deliver repeatable, production-grade operations.