diff --git a/cspell.json b/cspell.json index 0cc0c605..101b0628 100644 --- a/cspell.json +++ b/cspell.json @@ -8,6 +8,7 @@ "appname", "azurecontainerapps", "Benitez", + "BYOM", "candidateid", "candidatenumber", "chatbots", @@ -21,6 +22,7 @@ "devtunnel", "devtunnels", "docusign", + "DTMF", "Dunnam", "duplicative", "Durow", @@ -30,6 +32,7 @@ "Frontmatter", "Helpdesk", "hiringagent", + "HITL", "inlinehilite", "Jammes", "jobdescription", @@ -44,6 +47,7 @@ "mkdocs", "MSRC", "multimodal", + "Noda", "Omnichannel", "onboardsupport", "opensource", @@ -61,6 +65,7 @@ "resumenumber", "RXXXXX", "Signup", + "SSML", "Streamable", "subprocessor", "subprocessors", diff --git a/docs/commander-preview/01-advanced-mcp/index.md b/docs/commander-preview/01-advanced-mcp/index.md new file mode 100644 index 00000000..914548d0 --- /dev/null +++ b/docs/commander-preview/01-advanced-mcp/index.md @@ -0,0 +1,33 @@ +--- +prev: + text: 'Commander Preview Overview' + link: '/commander-preview' +next: + text: 'Computer Using Agents (CUA)' + link: '/commander-preview/02-computer-using-agents' +--- + +# ๐Ÿ”Œ Mission 01: Advanced MCP Server Usage + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Go beyond basic MCP concepts and connect your Copilot Studio agent to MCP servers using the MCP onboarding wizard or a custom connector in Power Apps. Configure authentication (API key or OAuth 2.0 with dynamic discovery, dynamic, or manual flows). Add MCP server tools and resources to your agent and selectively enable or disable individual tools. Explore the built-in MCP servers catalog - including Dataverse, Dynamics 365, Fabric, and the Microsoft 365 MCP servers - and troubleshoot common integration issues. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to connect your agent to an existing MCP server using the MCP onboarding wizard or a custom connector in Power Apps +1. How to configure MCP authentication (None, API key, OAuth 2.0 - dynamic discovery, dynamic, and manual) +1. How to add MCP server tools and resources to your agent and selectively enable or disable individual tools +1. How to use the built-in MCP servers catalog (Dataverse, Dynamics 365, and Fabric) +1. How to troubleshoot common MCP integration issues and known limitations + +## ๐Ÿ“– References + +- [Create a new MCP server](https://learn.microsoft.com/en-us/microsoft-copilot-studio/mcp-create-new-server) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/02-computer-using-agents/index.md b/docs/commander-preview/02-computer-using-agents/index.md new file mode 100644 index 00000000..648ef7df --- /dev/null +++ b/docs/commander-preview/02-computer-using-agents/index.md @@ -0,0 +1,33 @@ +--- +prev: + text: 'Advanced MCP Server Usage' + link: '/commander-preview/01-advanced-mcp' +next: + text: 'Code Interpreter' + link: '/commander-preview/03-code-interpreter' +--- + +# ๐Ÿ–ฅ๏ธ Mission 02: Computer Using Agents (CUA) + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Use Computer Use to let agents automate apps and websites with virtual mouse and keyboard. Learn when to use CUA vs RPA, how to prompt effectively, how to store secure credentials using Azure Key Vault, how to allow CUA to request human assistance (HITL), how to provide inputs, and the different types of hosts. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to enable and configure Computer Use for your agents +1. When to choose CUA over traditional RPA approaches +1. How to prompt computer-using agents effectively +1. How to store secure credentials with Azure Key Vault +1. How to implement human-in-the-loop assistance, provide inputs, and select appropriate host types + +## ๐Ÿ“– References + +- [Computer Use](https://learn.microsoft.com/en-us/microsoft-copilot-studio/computer-use) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/03-code-interpreter/index.md b/docs/commander-preview/03-code-interpreter/index.md new file mode 100644 index 00000000..8f3199bd --- /dev/null +++ b/docs/commander-preview/03-code-interpreter/index.md @@ -0,0 +1,33 @@ +--- +prev: + text: 'Computer Using Agents (CUA)' + link: '/commander-preview/02-computer-using-agents' +next: + text: 'Human in the Loop (HITL)' + link: '/commander-preview/04-human-in-the-loop' +--- + +# ๐Ÿ Mission 03: Code Interpreter + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Enable Code Interpreter to generate and execute Python code in a sandbox for data preparation, statistical analysis, table joins, forecasting, charting, and file generation using structured data (CSV, Excel). Leverage deterministic, reproducible computation instead of best-effort LLM reasoning. Ingest data via user-uploaded files in chat or SharePoint Documents library knowledge sources. Combine with image generation and custom skills to extend agent capabilities. Query and update Dataverse using code. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to enable Code Interpreter at the prompt and agent level +1. How to analyze structured data (CSV, Excel) for data preparation, statistical analysis, forecasting, and chart generation +1. The difference between deterministic (prompt-level) and non-deterministic (agent-level) execution +1. How to ingest structured data via user-uploaded files or SharePoint knowledge sources +1. How to extend capabilities with image generation, custom skills, and Dataverse code queries + +## ๐Ÿ“– References + +- [Code interpreter](https://learn.microsoft.com/en-us/microsoft-365-copilot/extensibility/code-interpreter) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/04-human-in-the-loop/index.md b/docs/commander-preview/04-human-in-the-loop/index.md new file mode 100644 index 00000000..854d9fa9 --- /dev/null +++ b/docs/commander-preview/04-human-in-the-loop/index.md @@ -0,0 +1,35 @@ +--- +prev: + text: 'Code Interpreter' + link: '/commander-preview/03-code-interpreter' +next: + text: 'Voice-enabled Agents' + link: '/commander-preview/05-voice-enabled-agents' +--- + +# ๐Ÿค Mission 04: Human in the Loop (HITL) + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Build multi-stage approval workflows in agent flows that combine AI-driven decisions with manual human review stages. Configure the Human in the Loop connector with manual and AI approval stages, conditions for dynamic routing, and the Request for Information (RFI) action to pause flows and collect structured input from reviewers. Map flow variables to approval inputs, write effective AI stage instructions, and review decisions in the approvals center. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to add and configure **multi-stage approvals** using the Human in the Loop connector in agent flows +1. How to set up **manual and AI approval stages** with approval types, assignees, input mappings, and automated document evaluation +1. How to add **conditions** to dynamically route, approve, reject, or skip stages based on business rules +1. How to use the **Request for Information** action to pause flows, collect structured input from reviewers, and map flow variables to approval inputs +1. How to write **effective AI stage instructions** and review AI decisions and rationale in the approvals center + +## ๐Ÿ“– References + +- [Agent flows overview](https://learn.microsoft.com/en-us/microsoft-copilot-studio/flows-overview) +- [Multi-stage AI Approvals](https://learn.microsoft.com/en-us/microsoft-copilot-studio/flows-advanced-approvals) +- [Request more information](https://learn.microsoft.com/en-us/microsoft-copilot-studio/flows-request-for-information) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/05-voice-enabled-agents/index.md b/docs/commander-preview/05-voice-enabled-agents/index.md new file mode 100644 index 00000000..bc30f961 --- /dev/null +++ b/docs/commander-preview/05-voice-enabled-agents/index.md @@ -0,0 +1,35 @@ +--- +prev: + text: 'Human in the Loop (HITL)' + link: '/commander-preview/04-human-in-the-loop' +next: + text: 'Localization' + link: '/commander-preview/06-localization' +--- + +# ๐ŸŽ™๏ธ Mission 05: Voice-enabled Agents + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Build agents with interactive voice response (IVR) capabilities using the Voice template. Handle phone calls with speech and dual-tone multi-frequency (DTMF) input. DTMF allows your agent users to use their phone keypad to select options or provide information. Configure barge-in so users can interrupt, set up silence detection and timeouts, customize your agent's voice with Speech Synthesis Markup Language (SSML), and manage call flow with call transfer and termination. Use Azure Communication Services to provision phone numbers for your agent. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to create a voice-enabled agent using the Voice template +1. How to collect user input via speech and dual-tone multi-frequency (DTMF) +1. How to configure barge-in, silence detection, timeouts, and customize voice using SSML +1. How to manage call flow with latency messages, call transfer, and termination +1. How to provision phone numbers with Azure Communication Services + +## ๐Ÿ“– References + +- [Voice overview](https://learn.microsoft.com/en-us/microsoft-copilot-studio/voice-overview) +- [Voice template](https://learn.microsoft.com/en-us/microsoft-copilot-studio/voice-build-from-template) +- [Configure voice capabilities](https://learn.microsoft.com/en-us/microsoft-copilot-studio/voice-configuration) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/06-localization/index.md b/docs/commander-preview/06-localization/index.md new file mode 100644 index 00000000..e3997da8 --- /dev/null +++ b/docs/commander-preview/06-localization/index.md @@ -0,0 +1,38 @@ +--- +prev: + text: 'Voice-enabled Agents' + link: '/commander-preview/05-voice-enabled-agents' +next: + text: 'Extend with Azure AI & BYOM' + link: '/commander-preview/07-azure-ai-byom' +--- + +# ๐ŸŒ Mission 06: Localization + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Configure multilingual agents that communicate in different languages and handle regional date and time correctly. Set primary and secondary languages, manage localization with JSON and ResX translation files, and enable automatic or dynamic language switching using generative orchestration. Localize Adaptive Cards with dynamic content, accommodate time zones using system variables and Power Fx, and test multilingual experiences in the test panel. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to configure primary and secondary languages with region settings +1. How to manage localization using JSON and ResX translation files +1. How to set up automatic language detection and dynamic language switching with generative orchestration +1. How to localize Adaptive Cards and test multilingual experiences in the test panel +1. How to accommodate time zones using `Conversation.LocalTimeZone`, Noda Time zone IDs, and Power Fx date conversion + +## ๐Ÿ“– References + +- [Configure and create multilingual agents](https://learn.microsoft.com/en-us/microsoft-copilot-studio/multilingual) +- [Language support](https://learn.microsoft.com/en-us/microsoft-copilot-studio/authoring-language-support) +- [Accommodate time zones](https://learn.microsoft.com/en-us/microsoft-copilot-studio/manage-date-and-time) +- [Regional settings and supported locales](https://learn.microsoft.com/en-us/microsoft-copilot-studio/data-localization) +- [Design effective language understanding - Localization and languages](https://learn.microsoft.com/en-us/microsoft-copilot-studio/guidance/language-understanding#localization-and-languages) +- [Auto-detect language sample solution](https://github.com/microsoft/CopilotStudioSamples/tree/main/AutoDetectLanguageSample) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/07-azure-ai-byom/index.md b/docs/commander-preview/07-azure-ai-byom/index.md new file mode 100644 index 00000000..2d0c5ff9 --- /dev/null +++ b/docs/commander-preview/07-azure-ai-byom/index.md @@ -0,0 +1,34 @@ +--- +prev: + text: 'Localization' + link: '/commander-preview/06-localization' +next: + text: 'Knowledge Deep Dive' + link: '/commander-preview/08-knowledge-deep-dive' +--- + +# ๐Ÿง  Mission 07: Extend with Azure AI & BYOM + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Use Azure AI Search for enterprise data retrieval, Azure AI Foundry for bring-your-own-data copilots, and BYOM to run custom or fine-tuned models. Add prompts as reusable tools inside topics to extend agent capabilities. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to integrate Azure AI Search for enterprise knowledge retrieval +1. How to use Azure AI Foundry for bring-your-own-data scenarios +1. How to bring your own models (BYOM) with custom or fine-tuned models +1. How to add prompts as reusable tools inside topics + +## ๐Ÿ“– References + +- [Azure AI Search](https://learn.microsoft.com/en-us/microsoft-copilot-studio/knowledge-azure-ai-search) +- [Azure AI Foundry](https://learn.microsoft.com/en-us/azure/ai-foundry/openai/concepts/use-your-data) +- [BYOM](https://learn.microsoft.com/en-us/ai-builder/byom-for-your-prompts) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/08-knowledge-deep-dive/index.md b/docs/commander-preview/08-knowledge-deep-dive/index.md new file mode 100644 index 00000000..4fd9aa31 --- /dev/null +++ b/docs/commander-preview/08-knowledge-deep-dive/index.md @@ -0,0 +1,34 @@ +--- +prev: + text: 'Extend with Azure AI & BYOM' + link: '/commander-preview/07-azure-ai-byom' +next: + text: 'Handoff to Human Agents' + link: '/commander-preview/09-handoff-to-human' +--- + +# ๐Ÿ“š Mission 08: Knowledge Deep Dive + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Optimize websites for agent ingestion, connect SharePoint and OneDrive content while understanding limitations and how your agent reasons over information. Use Bing Search for dynamic external knowledge retrieval. Understand security and identity topics on how information is accessed and used. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to optimize public websites for agent knowledge ingestion +1. How to connect SharePoint and OneDrive content as knowledge sources +1. How agents reason over connected knowledge and current limitations +1. How to use Bing Search for dynamic external knowledge retrieval +1. How security and identity govern knowledge access and usage + +## ๐Ÿ“– References + +- [Knowledge sources](https://learn.microsoft.com/en-us/microsoft-copilot-studio/knowledge-copilot-studio) +- [Guidance for public websites](https://learn.microsoft.com/en-us/microsoft-copilot-studio/guidance/generative-ai-public-websites) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/09-handoff-to-human/index.md b/docs/commander-preview/09-handoff-to-human/index.md new file mode 100644 index 00000000..38d3e50e --- /dev/null +++ b/docs/commander-preview/09-handoff-to-human/index.md @@ -0,0 +1,36 @@ +--- +prev: + text: 'Knowledge Deep Dive' + link: '/commander-preview/08-knowledge-deep-dive' +next: + text: 'Governance with Auth, Channels & DLP' + link: '/commander-preview/10-governance-auth-dlp' +--- + +# ๐Ÿ”„ Mission 09: Handoff to Human Agents + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Configure your agent to hand off conversations to live human agents seamlessly and contextually. Understand how the Escalate system topic and Transfer conversation node work, and how handoff can be triggered implicitly (when a user asks for a human) or explicitly (within a topic flow). Learn how context variables such as conversation history, topics, and user-defined variables are passed to the engagement hub so the live agent can resume without the customer repeating themselves. Explore supported engagement hubs including Dynamics 365 Customer Service, ServiceNow, Salesforce, LivePerson, and generic hubs via custom adapters. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to configure the Escalate system topic and add a Transfer conversation node +1. The difference between implicit triggers (user requests a human) and explicit triggers (topic-driven escalation) +1. How context variables (conversation history, topics, phrases, private messages) are passed to the engagement hub +1. How to connect to supported engagement hubs: Dynamics 365 Customer Service, ServiceNow, Salesforce, LivePerson, and generic hubs +1. How to customize agent behavior for escalation scenarios (greeting, escalate link, no-match messages) + +## ๐Ÿ“– References + +- [Customer engagement overview](https://learn.microsoft.com/en-us/microsoft-copilot-studio/customer-copilot-overview) +- [Handoff pattern](https://learn.microsoft.com/en-us/microsoft-copilot-studio/advanced-hand-off) +- [Generic handoff configuration](https://learn.microsoft.com/en-us/microsoft-copilot-studio/configure-generic-handoff) +- [Configure handoff to Dynamics 365 Customer Service](https://learn.microsoft.com/en-us/microsoft-copilot-studio/configuration-hand-off-omnichannel) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/10-governance-auth-dlp/index.md b/docs/commander-preview/10-governance-auth-dlp/index.md new file mode 100644 index 00000000..8e129042 --- /dev/null +++ b/docs/commander-preview/10-governance-auth-dlp/index.md @@ -0,0 +1,35 @@ +--- +prev: + text: 'Handoff to Human Agents' + link: '/commander-preview/09-handoff-to-human' +next: + text: 'Test, Monitor, Debug & Analytics' + link: '/commander-preview/11-test-monitor-debug' +--- + +# ๐Ÿ” Mission 10: Governance Essentials with Authentication, Channels & DLP + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Secure agent access using Single Sign-On with Entra ID and manual authentication for APIs. Compare Streaming API vs Direct Line for embedding agents in custom apps. Prevent anonymous access using DLP policies. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to configure Single Sign-On (SSO) with Entra ID +1. How to set up manual authentication for API access +1. How to compare Streaming API vs Direct Line for custom app embedding +1. How to prevent anonymous access using DLP policies + +## ๐Ÿ“– References + +- [Single Sign-On](https://learn.microsoft.com/en-us/microsoft-copilot-studio/configure-sso) +- [Manual authentication](https://learn.microsoft.com/en-us/microsoft-copilot-studio/configuration-end-user-authentication) +- [Data loss prevention policies](https://learn.microsoft.com/en-us/microsoft-copilot-studio/admin-data-loss-prevention) +- [Connect agent to Teams and Microsoft 365](https://learn.microsoft.com/en-us/microsoft-copilot-studio/publication-add-bot-to-microsoft-teams) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/11-test-monitor-debug/index.md b/docs/commander-preview/11-test-monitor-debug/index.md new file mode 100644 index 00000000..1663933c --- /dev/null +++ b/docs/commander-preview/11-test-monitor-debug/index.md @@ -0,0 +1,40 @@ +--- +prev: + text: 'Governance with Auth, Channels & DLP' + link: '/commander-preview/10-governance-auth-dlp' +next: + text: 'Deploy Agents with Solutions & Pipelines' + link: '/commander-preview/12-deploy-solutions-pipelines' +--- + +# ๐Ÿงช Mission 11: Test, Monitor, Debug & Analytics + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Use the built-in Test panel to walk through conversations, track between topics, and read the real-time Activity map during generative orchestration. Inspect variable values, save conversation snapshots for diagnostics, and flag issues for Microsoft support. Go beyond manual testing with agent evaluation: create test sets, run automated evaluations with user profiles, and compare scores across agent changes. Monitor production performance using Analytics dashboards for KPIs like session outcomes, CSAT scores, and topic trends. Configure custom telemetry with Azure Application Insights and build Power BI dashboards for advanced reporting. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to use the Test panel, track between topics, and read the real-time Activity map +1. How to inspect variables, save conversation snapshots, and flag issues for support +1. How to create test sets, configure user profiles, and run automated agent evaluations +1. How to interpret evaluation scores, drill into results, and compare runs across agent changes +1. How to monitor production KPIs with Analytics dashboards and configure custom telemetry with Application Insights and Power BI + +## ๐Ÿ“– References + +- [Test your agent](https://learn.microsoft.com/en-us/microsoft-copilot-studio/authoring-test-bot) +- [Agent evaluation overview](https://learn.microsoft.com/en-us/microsoft-copilot-studio/analytics-agent-evaluation-intro) +- [Create test cases for evaluation](https://learn.microsoft.com/en-us/microsoft-copilot-studio/analytics-agent-evaluation-create) +- [Evaluation test methods](https://learn.microsoft.com/en-us/microsoft-copilot-studio/analytics-agent-evaluation-overview) +- [Review evaluation results](https://learn.microsoft.com/en-us/microsoft-copilot-studio/analytics-agent-evaluation-results) +- [Activity maps](https://learn.microsoft.com/en-us/microsoft-copilot-studio/authoring-review-activity) +- [Analytics overview](https://learn.microsoft.com/en-us/microsoft-copilot-studio/analytics-overview) +- [Capture telemetry with Application Insights](https://learn.microsoft.com/en-us/microsoft-copilot-studio/advanced-bot-framework-composer-capture-telemetry) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/12-deploy-solutions-pipelines/index.md b/docs/commander-preview/12-deploy-solutions-pipelines/index.md new file mode 100644 index 00000000..5a746d01 --- /dev/null +++ b/docs/commander-preview/12-deploy-solutions-pipelines/index.md @@ -0,0 +1,40 @@ +--- +prev: + text: 'Test, Monitor, Debug & Analytics' + link: '/commander-preview/11-test-monitor-debug' +next: + text: 'Next Steps and Best Practices' + link: '/commander-preview/13-next-steps-best-practices' +--- + +# ๐Ÿš€ Mission 12: Deploy Agents with Solutions & Pipelines + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Plan and execute enterprise application lifecycle management (ALM) for your agents. Understand environment strategies for dev, test, and production, and package agents in custom solutions with managed and unmanaged layers. Export and import solutions across environments, create reusable component collections, and configure agent connections with On-Behalf-Of (OBO) authentication. Control sharing with Editor and Viewer roles and deploy at scale using Power Platform Pipelines. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to plan environment strategies for dev, test, and production, including trial-to-production conversion and backup/restore operations +1. How to create custom solutions, manage solution layers (managed vs unmanaged), and export/import with required objects +1. How to create and share reusable component collections with dependency management +1. How to configure agent connections, On-Behalf-Of (OBO) authentication, and sharing with Editor and Viewer roles +1. How to deploy solutions at scale using Power Platform Pipelines + +## ๐Ÿ“– References + +- [Environments overview](https://learn.microsoft.com/en-us/microsoft-copilot-studio/environments-first-run-experience) +- [Solutions overview](https://learn.microsoft.com/en-us/microsoft-copilot-studio/authoring-solutions-overview) +- [Export and import agents](https://learn.microsoft.com/en-us/microsoft-copilot-studio/authoring-solutions-import-export) +- [Export and import agent components](https://learn.microsoft.com/en-us/microsoft-copilot-studio/authoring-export-import-copilot-components) +- [Manage connections](https://learn.microsoft.com/en-us/microsoft-copilot-studio/authoring-connections) +- [Sharing controls and limits](https://learn.microsoft.com/en-us/microsoft-copilot-studio/admin-sharing-controls-limits) +- [Solution management](https://learn.microsoft.com/en-us/power-platform/release-plan/2024wave2/microsoft-copilot-studio/solution-management-copilot-studio) +- [Power Platform Pipelines](https://learn.microsoft.com/en-us/power-platform/alm/pipelines) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/13-next-steps-best-practices/index.md b/docs/commander-preview/13-next-steps-best-practices/index.md new file mode 100644 index 00000000..e48b41bb --- /dev/null +++ b/docs/commander-preview/13-next-steps-best-practices/index.md @@ -0,0 +1,30 @@ +--- +prev: + text: 'Deploy Agents with Solutions & Pipelines' + link: '/commander-preview/12-deploy-solutions-pipelines' +--- + +# ๐Ÿ Mission 13: Next Steps and Best Practices + +## ๐Ÿ•ต๏ธโ€โ™‚๏ธ **CLASSIFIED** + +## ๐ŸŽฏ Mission Brief + +Apply the Copilot Studio Implementation Guide for real-world deployments. Use Power CAT tools, monitoring dashboards, and the Well-Architected Framework for long-term governance and operational excellence. + +## ๐Ÿ”Ž Objectives + +In this mission, you'll learn: + +1. How to apply the Copilot Studio Implementation Guide +1. How to leverage Power CAT tools for agent management +1. How to set up monitoring and governance dashboards +1. How to use the Well-Architected Framework for long-term operational excellence + +## ๐Ÿ“– References + +- [Copilot Studio Implementation Guide](https://www.microsoft.com/en-us/microsoft-copilot/blog/copilot-studio/new-microsoft-copilot-studio-implementation-guide/) +- [Well-Architected Framework](https://learn.microsoft.com/en-us/power-platform/well-architected/operational-excellence/tools-processes) + +> [!NOTE] +> ๐Ÿšง This mission is under construction. Check back soon for the full walkthrough. diff --git a/docs/commander-preview/index.md b/docs/commander-preview/index.md new file mode 100644 index 00000000..fbf88301 --- /dev/null +++ b/docs/commander-preview/index.md @@ -0,0 +1,51 @@ +--- +next: + text: 'Advanced MCP Server Usage' + link: '/commander-preview/01-advanced-mcp' +lastUpdated: false +--- + +# Welcome Commander (Preview) + +![Copilot Studio Agent Academy Commander](../images/mcs-agent-academy-commander-badge.png) + +> [!NOTE] +> โš ๏ธ This is a **preview** of the Commander curriculum. Content is under active development. + +**Welcome, Commander.** +Your elite mission, should you choose to accept it, is to master the most advanced capabilities of **Microsoft Copilot Studio** and lead the charge in building **production-ready, enterprise-scale agent solutions**. + +This advanced training takes you beyond multi-agent orchestration into the cutting edge of agent development: from computer-using agents and code interpretation to voice-enabled experiences, governance, and deployment at scale. + +## ๐Ÿงฌ Who This Is For + +This elite course is ideal for: + +- **Solution architects** designing enterprise-scale AI deployments +- **Developers** building production-ready, governed agent solutions +- **IT professionals** implementing advanced security and ALM practices +- **AI specialists** extending Copilot Studio with Azure AI and custom models +- Anyone ready to **command** the full breadth of Copilot Studio capabilities + +## ๐Ÿงญ Curriculum Overview + +This academy is structured as a progressive series of advanced operations. Each mission deepens your mastery of enterprise agent capabilities. + +| Mission | Title | Operation Briefing | +| --------- | ------- | ------------------- | +| `01` | ๐Ÿ”Œ [Advanced MCP Server Usage](./01-advanced-mcp/index.md) | Connect to Dataverse MCP, add existing MCP servers, and explore the Power Apps MCP server | +| `02` | ๐Ÿ–ฅ๏ธ [Computer Using Agents (CUA)](./02-computer-using-agents/index.md) | Automate apps and websites with virtual mouse and keyboard interactions | +| `03` | ๐Ÿ [Code Interpreter](./03-code-interpreter/index.md) | Run Python scripts in a sandbox for analysis, charting, and file generation | +| `04` | ๐Ÿค [Human in the Loop (HITL)](./04-human-in-the-loop/index.md) | Create multi-stage approval flows mixing AI decisions and manual review | +| `05` | ๐ŸŽ™๏ธ [Voice-enabled Agents](./05-voice-enabled-agents/index.md) | Build agents that handle phone calls, speech, and dual-tone multi-frequency (DTMF) input | +| `06` | ๐ŸŒ [Localization](./06-localization/index.md) | Build multilingual agents with language detection, dynamic switching, and time zone accommodation | +| `07` | ๐Ÿง  [Extend with Azure AI & BYOM](./07-azure-ai-byom/index.md) | Integrate Azure AI Search, Azure AI Foundry, and bring your own models | +| `08` | ๐Ÿ“š [Knowledge Deep Dive](./08-knowledge-deep-dive/index.md) | Optimize knowledge sources, connect SharePoint, and understand security | +| `09` | ๐Ÿ”„ [Handoff to Human Agents](./09-handoff-to-human/index.md) | Configure escalation to live agents with context transfer across supported engagement hubs | +| `10` | ๐Ÿ” [Governance Essentials with Authentication, Channels & DLP](./10-governance-auth-dlp/index.md) | Secure agent access with SSO, authentication, and DLP policies | +| `11` | ๐Ÿงช [Test, Monitor, Debug & Analytics](./11-test-monitor-debug/index.md) | Test panel, activity maps, conversation snapshots, automated agent evaluations, Analytics dashboards, custom telemetry, and Power BI reporting | +| `12` | ๐Ÿš€ [Deploy Agents with Solutions & Pipelines](./12-deploy-solutions-pipelines/index.md) | Environment strategies, solutions, component collections, connections, sharing controls, and Pipelines | +| `13` | ๐Ÿ [Next Steps and Best Practices](./13-next-steps-best-practices/index.md) | Apply implementation guides, Power CAT tools, and the Well-Architected Framework | + + +Analytics