> For the complete documentation index, see [llms.txt](https://dyns.ntd.asia/power-dynamics/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio.md).

# Study Copilot Studio

- [Mission](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/mission.md): The learning plan for the Copilot Studio course — mission, sources, and full session map. Course complete as of Session 8.3 (Groups 1–8).
- [Foundations](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/foundations.md)
- [What's Copilot Studio?](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/foundations/whats-copilot-studio.md): Session 1.1 — what the platform actually is, the three harnesses underneath it, and how they map to the knowledge-vs-skill frame from "What Is AI, Really?"
- [How an Agent Decides](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/foundations/how-an-agent-decides.md): Session 1.2 — classic vs. generative orchestration, the three control layers for agent autonomy, and why the GitHub Copilot harness skips the toggle entirely.
- [Core Agent Building Blocks](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/core-agent-building-blocks.md)
- [Topics & the Authoring Canvas](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/core-agent-building-blocks/topics-and-the-authoring-canvas.md): Session 2.1 — the authoring canvas, topics, trigger phrases, and the core node types, with a Northwind Outfitters Returns & Exchanges topic built step by step.
- [Entities and Variables](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/core-agent-building-blocks/entities-and-variables.md): Session 2.2 — entities, custom closed-list and regex entities, slot filling (including proactive slot filling), and variable scope (topic vs. global), built on the Northwind Outfitters Returns & Excha
- [Generative Answers and Instructions](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/core-agent-building-blocks/generative-answers-and-instructions.md): Session 2.3 of 4 — Core Agent Building Blocks. The generative answers node and agent instructions in Microsoft Copilot Studio, verified against Microsoft Learn.
- [Review & Retrieval](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/core-agent-building-blocks/review-and-retrieval.md): Session 2.4 of 4 — Group 2, Core Agent Building Blocks. Combines topics, entities, variables, and generative answers into one build; adds pass/return variables between topics, one-of-multiple-entities
- [Knowledge & Grounding](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/knowledge-and-grounding.md): Group 3 of the Copilot Studio study plan — knowledge sources, grounding mechanics, and Azure AI Search.
- [Knowledge Sources: Files, SharePoint, and the Web](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/knowledge-and-grounding/knowledge-sources-files-sharepoint-and-the-web.md): Session 3.1 of 3 — Group 3: Knowledge & Grounding. Knowledge source fundamentals: files, SharePoint, and public websites.
- [Grounding Mechanics and Citations](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/knowledge-and-grounding/grounding-mechanics-and-citations.md): Session 3.2 of the Copilot Studio course — Group 3: Knowledge & Grounding. How allow-ungrounded-responses, citation withholding, official sources, and per-channel citation limits actually behave at an
- [Azure AI Search & Custom Knowledge Sources](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/knowledge-and-grounding/azure-ai-search-and-custom-knowledge-sources.md): Session 3.3 of the Copilot Studio course — Group 3: Knowledge & Grounding (final session). Connecting a bring-your-own Azure AI Search index, and the two competing paths for custom API knowledge sourc
- [Tools & Actions](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/tools-and-actions.md): Group 4 of the Copilot Studio study plan — connectors, agent flows, prompts, REST APIs, and adaptive cards: what the agent can actually do, not just say.
- [Connectors and Power Platform Actions](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/tools-and-actions/connectors-and-power-platform-actions.md): Session 4.1 of the Copilot Studio course — Group 4: Tools & Actions (session 1 of 5). The first mechanism for giving an agent something to do, not just something to say: Power Platform connectors.
- [Agent Flows: Human-in-the-Loop and Error Handling](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/tools-and-actions/agent-flows-human-in-the-loop-and-error-handling.md): Session 4.2 of 5 — Group 4: Tools & Actions. Request information, multistage/AI approvals, asynchronous responses, and three layers of agent-flow error handling.
- [Prompts and REST APIs: Off the Connector Path](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/tools-and-actions/prompts-and-rest-apis-off-the-connector-path.md): Session 4.3 of 5 — Group 4: Tools & Actions. The prompt tool for single-turn model tasks, plus two ways to call a raw REST API: the topic-level HTTP Request node and the preview agent-level REST API t
- [Adaptive Cards and Response Formatting](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/tools-and-actions/adaptive-cards-and-response-formatting.md): Session 4.4 of Group 4 — Tools & Actions. Fourth of five sessions, after connectors (4.1), agent flows (4.2), and prompts/REST APIs (4.3). 4.5's closing Northwind build is next.
- [Review & Build: Northwind Tool Chain](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/tools-and-actions/review-and-build-northwind-tool-chain.md): Session 4.5 of Group 4 — Tools & Actions. Fifth and last session, closing the group by wiring 4.3's REST API tool, 4.2's human-review flow, and 4.4's adaptive-card completion into one Northwind build.
- [Agentic Behavior: Autonomous Agents](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/agentic-behavior-autonomous-agents.md): Group 5 - an agent that starts itself: triggers, decision boundaries, and guardrails.
- [Triggers, Boundaries, and Guardrails](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/agentic-behavior-autonomous-agents/triggers-boundaries-and-guardrails.md): Session 5.1 of Group 5 - Agentic Behavior. First session, opening the group: how an autonomous agent starts itself, what it's allowed to do, and how you keep it inside the lines.
- [Build & Deploy: The Return Follow-Up Trigger](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/agentic-behavior-autonomous-agents/build-and-deploy-the-return-follow-up-trigger.md): Session 5.2 of Group 5 - Agentic Behavior: Autonomous Agents. Builds, tests, and publishes 5.1's Northwind return-follow-up trigger.
- [Monitoring and Governing Autonomy](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/agentic-behavior-autonomous-agents/monitoring-and-governing-autonomy.md): Session 5.3 of Group 5 - Agentic Behavior: Autonomous Agents. The Activity page, Monitor page, and Purview audit logs - three tools, three clocks, for watching the 5.2 Northwind trigger.
- [Multi-Agent Orchestration](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/multi-agent-orchestration.md): Group 6 - delegating to another agent instead of building one that does everything: child agents, connected agents, Foundry, MCP-connected tools, A2A, and Fabric data agents.
- [Multi-Agent Design Patterns and Best Practices](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/multi-agent-orchestration/multi-agent-design-patterns-and-best-practices.md): Session 6.1 of Group 6 - Multi-Agent Orchestration: child vs. connected agents, when a subtask earns its own agent, and the nine disciplines behind a working multi-agent build.
- [Building Child and Connected Agents](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/multi-agent-orchestration/building-child-and-connected-agents.md): Session 6.2 of Group 6: the actual build steps for a child agent and a connected agent, then wiring Northwind's Billing/Product-Support split into two published connected agents.
- [Integrating a Foundry Agent](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/multi-agent-orchestration/integrating-a-foundry-agent.md): Session 6.3 of Group 6: connecting a Microsoft Foundry agent (preview) to Copilot Studio — the first agent type that lives outside Copilot Studio entirely.
- [Agent2Agent Protocol and Fabric Data Agents](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/multi-agent-orchestration/agent2agent-protocol-and-fabric-data-agents.md): Session 6.4 of Group 6 - Multi-Agent Orchestration: A2A protocol and the two harness-dependent ways to add a Fabric data agent.
- [Review & Build: Northwind Multi-Agent Handoff](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/multi-agent-orchestration/review-and-build-northwind-multi-agent-handoff.md): Session 6.5 of Group 6 - Multi-Agent Orchestration: wiring all five connection types into one live Northwind orchestrator and testing the seams between them.
- [Enterprise Integration & Governance](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/enterprise-integration-and-governance.md): Group 7: identity, security, responsible AI, and ALM for enterprise Copilot Studio deployments.
- [Identity and Security](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/enterprise-integration-and-governance/identity-and-security.md): Session 7.1: agent-level authentication, the security scan, Entra Agent ID, and how far a DLP data policy's reach extends.
- [Responsible AI: Principles, Safety Filters, and Human Oversight](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/enterprise-integration-and-governance/responsible-ai-principles-safety-filters-and-human-oversight.md): Copilot Studio Session 7.2: the four-vs-six responsible AI principles conflict, content moderation levels, ContentFiltered, and human oversight for prompts.
- [ALM Foundations: Solutions and Environment Variables](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/enterprise-integration-and-governance/alm-foundations-solutions-and-environment-variables.md): Copilot Studio Session 7.3: solutions as ALM containers, managed vs. unmanaged, environment strategy, and environment variables.
- [ALM Automation: Pipelines and CI/CD](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/enterprise-integration-and-governance/alm-automation-pipelines-and-ci-cd.md): Copilot Studio Session 7.4: Power Platform pipelines, Azure DevOps Build Tools, and GitHub Actions compared, plus the native Git vs. VS Code pull-request review gap.
- [Testing & Capstone](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/testing-and-capstone.md): Group 8: proving an agent works with a number instead of a feeling, then building one system that pulls every prior group together.
- [Evaluating Agents: Test Sets and Evaluation Methods](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/testing-and-capstone/evaluating-agents-test-sets-and-evaluation-methods.md): Copilot Studio Session 8.1: test sets, all seven evaluation methods, and reading results — applied to Northwind.
- [Conversational Evaluation, Automation & the GitHub Copilot Harness](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/testing-and-capstone/conversational-evaluation-automation-and-the-github-copilot-harness.md): Session 8.2: multi-turn test sets, REST API/connector automation, the GitHub Copilot harness's Evaluate preview, and what the Copilot Agent Kit's pipeline test gate actually is.
- [Capstone: The Full Northwind Outfitters Agent](https://dyns.ntd.asia/power-dynamics/artificial-intelligence/study-copilot-studio/testing-and-capstone/capstone-the-full-northwind-outfitters-agent.md): Session 8.3 (capstone, Group 8 complete): the full Northwind Outfitters agent — conversational, autonomous, and multi-agent — run together for the first time, with three flagged gaps re-checked live.