Merge pull request #314061 from dm-chelupati/docs-agent/learn-2026-03-31

[Docs] SRE Agent Learn update — 2026-03-21

Author: denrea

articles/sre-agent/connectors.md

@@ -3,7 +3,7 @@ title: Connectors in Azure SRE Agent
 description: Extend your agent's capabilities to external data sources, collaboration tools, and custom APIs using connectors.
 ms.topic: concept-article
 ms.service: azure-sre-agent
-ms.date: 03/18/2026
+ms.date: 03/25/2026
 author: craigshoemaker
 ms.author: cshoe
 ms.ai-usage: ai-assisted
@@ -128,6 +128,52 @@ The following table describes scenarios where automatic recovery can't resolve t
 
 When your agent tries to use a tool from a disconnected MCP server, it provides a clear error explaining what went wrong (for example, "Connection is disconnected and reconnection failed."). Your agent doesn't silently fail or produce incorrect results.
 
+## Browse and manage connectors
+
+Open the Connectors page (**Builder** > **Connectors**) to see your connectors organized into collapsible category groups. All groups are expanded by default.
+
+| Category | What it includes |
+|---|---|
+| **Notification** | Teams and Outlook messaging connectors |
+| **Telemetry** | Azure Data Explorer, Datadog, Dynatrace, Elasticsearch, New Relic, Splunk, and other monitoring connectors |
+| **Code Repository** | GitHub OAuth, Azure DevOps OAuth, and documentation connectors |
+| **MCP** | GitHub MCP, generic MCP servers, and custom MCP integrations |
+| **Incidents** | Incident management connectors |
+| **Deployment** | Deployment pipeline connectors (EV2) |
+| **Other** | Connectors that don't fit other categories |
+
+The **Code Repository** category includes nested sub-groups that organize your repositories by provider and organization. GitHub connectors appear under a **GitHub** sub-group, and Azure DevOps connectors are grouped by their ADO organization (for example, **contoso (Azure DevOps)**). Each sub-group has its own expand/collapse control and count badge. Sub-groups appear automatically when you have connectors from two or more distinct providers or organizations.
+
+Each category header shows the number of connectors in that group. When you collapse a category, a red badge appears if any connector in that group has a connection issue, so you can spot problems at a glance without expanding every section.
+
+Use the toolbar controls to manage your view:
+
+- **Expand all / Collapse all**: Toggle all category groups at once
+- **Category filter**: Show only connectors in a specific category
+- **Search**: Find connectors by name (switches to a flat list for keyword lookup)
+
+Only categories that contain at least one connector are displayed. When you search for a connector by name, the page switches to a flat list view for faster filtering.
+
+### Add a connector
+
+Select **Add connector** to open the connector wizard. The first step presents connectors organized by tab:
+
+| Tab | What it shows |
+|---|---|
+| **Telemetry** | Azure Data Explorer, Datadog, Dynatrace, Elasticsearch, New Relic, Splunk, and other monitoring connectors |
+| **Notification** | Outlook and Teams connectors |
+| **Code Repository** | GitHub OAuth and Azure DevOps OAuth connectors |
+| **MCP** | GitHub MCP, generic MCP server |
+| **Incidents** | Incident management connectors |
+| **Deployment** | Deployment pipeline connectors (EV2) |
+
+Select a connector card, then follow the setup steps for that connector type. Use the search box to find a connector by name across all tabs.
+
+> [!NOTE]
+> **Repos moved to Knowledge Sources**
+>
+> Source code repository management moved to **Knowledge Sources**. Navigate to **Builder** > **Knowledge sources** to manage your connected repositories.
+
 ## Who can configure connectors
 
 Connector management requires **write** permission on the agent. The following table shows which roles can configure connectors.

articles/sre-agent/favorites-mine-filter.md

@@ -17,7 +17,7 @@ Use **Favorites** and the **Mine** filter to organize your agent's chat sidebar.
 
 > [!TIP]
 > - Star any thread to pin it in a dedicated **Favorites** section at the top of your sidebar.
-> - Use the **Mine** filter to see only threads you created or favorited.
+> - The **Mine** filter is active by default—your sidebar shows only your threads when you open it.
 > - Favorites persist across sessions and are stored per-user in the backend, not just in your browser.
 > - Works immediately on any agent requiring no setup or configuration needed.
 
@@ -39,15 +39,15 @@ Use the following steps to organize your chat sidebar with favorites and the **M
 
 :::image type="content" source="media/favorites-mine-filter/favorites-sidebar-categories.png" alt-text="Screenshot of chat sidebar showing Favorites section at the top and Chats section below." lightbox="media/favorites-mine-filter/favorites-sidebar-categories.png":::
 
-**3. Use Mine to focus.** Select the **Mine** toggle button above the thread list to filter both sections. In **Favorites**, the filter shows threads you personally starred. In **Chats**, the filter shows threads you created. One select removes the noise from shared agents.
+**3. Mine is on by default.** The **Mine** filter is active when you open the sidebar, showing only threads you started in **Chats** and threads you starred in **Favorites**. Toggle it off to see all threads on the agent.
 
 :::image type="content" source="media/favorites-mine-filter/favorites-mine-filter-active.png" alt-text="Screenshot of Mine filter button active with filtered thread list." lightbox="media/favorites-mine-filter/favorites-mine-filter-active.png":::
 
 **4. Combine filters for precision.** **Mine** and **Unread** are independent toggles. Activate both to see only your unread threads across both categories.
 
 **5. Unfavorite when done.** Right-click a favorited thread and select **Remove from favorites** to move it back to the **Chats** section. Keep your **Favorites** list focused on active work.
 
-**6. Favorites persist across sessions.** The backend stores your favorites per user. They survive page refreshes, browser restarts, and device switches. Filter toggles (**Mine**, **Unread**) reset on page refresh.
+**6. Favorites persist across sessions.** The backend stores your favorites per user. They survive page refreshes, browser restarts, and device switches. The **Mine** filter is active by default—when you open the sidebar, you see your own threads first. The **Unread** filter resets on page refresh.
 
 ## What makes this approach different
 

articles/sre-agent/model-provider-selection.md

@@ -3,36 +3,45 @@ title: Model Provider Selection in Azure SRE Agent
 description: Learn how to choose your Azure SRE Agent AI model provider and model to match your operational workloads and data residency requirements.
 ms.topic: concept-article
 ms.service: azure-sre-agent
-ms.date: 03/30/2026
+ms.date: 04/02/2026
 author: dm-chelupati
 ms.author: dchelupati
 ms.ai-usage: ai-assisted
-ms.custom: model provider, model selection, Claude, GPT-5, Anthropic, Azure OpenAI, data residency, EUDB
+ms.custom: model provider, Anthropic, Azure OpenAI, data residency, EUDB
 #customer intent: As an SRE, I want to choose the right AI model provider for my agent so that I can balance performance, cost, and data residency requirements.
 ---
 
 # Model provider selection in Azure SRE Agent
 
-Choose between Azure OpenAI (GPT-5 family) and Anthropic (Claude family) as your agent's AI provider. Change models anytime in **Settings** > **Basics**—no downtime, takes effect on the next conversation.
+Choose between Azure OpenAI (GPT-5 family) and Anthropic (Claude family) as your agent's AI provider. Change providers anytime in **Settings** > **Basics**—no downtime, takes effect on the next conversation.
 
 > [!TIP]
 > - Choose between **Azure OpenAI** (GPT-5 family) and **Anthropic** (Claude family) as your agent's AI provider
+> - Your agent automatically selects the best model within your chosen provider—no manual model configuration needed
 > - Default provider depends on region: **Anthropic** in East US 2 and Australia East, **Azure OpenAI** in Sweden Central (EU data residency)
-> - Change models anytime in **Settings > Basics**—no downtime, takes effect on the next conversation
+> - Change providers anytime in **Settings > Basics**—no downtime, takes effect on the next conversation
+
+## Available providers
+
+| Provider | Models included | Strengths |
+|----------|----------------|----------|
+| **Azure OpenAI** | GPT-5 (default), GPT-5.2 | Structured output, function calling, fast tool execution |
+| **Anthropic** | Claude Opus 4.5, Claude Sonnet 4.5, Claude Sonnet 4.6, Claude Haiku 4.5 | Long-context reasoning, code analysis, built-in web search |
+
+Your agent automatically selects the best model within your chosen provider for each task. You select the provider; the agent handles model routing.
 
 ## How to change your model provider
 
 1. Go to **Settings > Basics** in the agent portal.
 2. Scroll to the **Model provider** section.
 3. Select a provider from the **Provider** dropdown.
-4. Select a specific model from the **Model** dropdown, or leave it as **Automatic** to let the agent choose the best model for each task.
-5. Select **Save**.
+4. Select **Save**.
 
-The change takes effect immediately—your next conversation uses the new model. No restart is required.
+The change takes effect immediately—your next conversation uses the new provider. No restart is required. Your agent automatically selects the best model within the provider for each task.
 
 ### Pricing
 
-Different models have different AAU rates. See [Pricing and billing—AAU rates by model](pricing-billing.md#aau-rates-by-model) for the current rates per token type.
+Different models have different AAU rates. Select the ℹ icon next to the **Model provider** label to open the pricing page, or see [Pricing and billing—AAU rates by model](pricing-billing.md#aau-rates-by-model) for the current rates per token type.
 
 ### Data residency
 
@@ -50,19 +59,20 @@ If EU data residency is a requirement for your organization, select Azure OpenAI
 
 ## What makes this approach different
 
-**Flexible model routing**—Your agent automatically routes different task types (reasoning, code generation, quick responses) to the appropriate model within your selected provider. You choose the provider; the agent optimizes model usage per task.
+**Automatic model routing**—Your agent automatically selects the best model within your chosen provider for each task. Reasoning, code generation, and quick responses each use the optimal model—no manual configuration needed.
 
-**Capability unlocks**—Selecting a Claude model automatically enables web search. Your agent can search the internet during investigations without any extra configuration.
+**Capability unlocks**—Selecting Anthropic as your provider automatically enables web search. Your agent can search the internet during investigations without any extra configuration.
 
-**No migration required**—Changing models is a settings change. Existing knowledge, memory, connectors, and scheduled tasks all work with any model provider.
+**No migration required**—Changing providers is a settings change. Existing knowledge, memory, connectors, and scheduled tasks all work with any model provider.
 
 ## Before and after
 
 | Aspect | Before | After |
 |--------|--------|-------|
-| **Model choice** | Single default model | Multiple providers and models to choose from |
-| **Web search** | Not available | Automatic with Claude models |
-| **Switching cost** | N/A | Zero—change in Settings, effective immediately |
+| **Provider choice** | Single default provider | Choose between Azure OpenAI and Anthropic |
+| **Model routing** | Fixed model | Agent automatically selects the best model per task |
+| **Web search** | Not available | Automatic with Anthropic provider |
+| **Switching cost** | N/A | Zero—change provider in Settings, effective immediately |
 
 ## Related content
 

articles/sre-agent/threads.md

@@ -53,6 +53,8 @@ Use the two filter buttons at the top of the thread list to find relevant conver
 | **Unread** | Only threads you didn't read yet. |
 | **Mine** | Only threads you started (excludes scheduled tasks and agent-initiated threads). |
 
+The **Mine** filter is active by default. When you open the sidebar, you see your own threads first. Toggle it off to see all threads on the agent.
+
 When active, a filter button appears filled (blue). You can activate both filters at the same time so that only threads matching **both** conditions appear.
 
 :::image type="content" source="media/threads/thread-filter-mine-active.png" alt-text="Screenshot of thread sidebar with the Mine filter active, showing only user-initiated threads.":::