If you’re setting your sights on the Microsoft AI Agent Builder Associate (AB-620) certification, you’re likely already aware of its focus—Copilot Studio. This exam zeroes in on building scalable AI agents pivotal to enterprise AI development. Having recently taken the exam myself, I want to distill what I’ve learned into a guide that can help you pass with ease.

For those gearing up for the AB-620, I’ve compiled a comprehensive study guide emphasizing the essential concepts you need to tackle: from Microsoft Copilot Studio and agent orchestration to enterprise integrations and multi-agent architectures. Our journey will navigate through Model Context Protocol (MCP), Agent2Agent (A2A), adaptive cards, and more. This preparatory journey isn’t just a rote memorization of facts—it aims to impart an understanding of underlying principles and technologies, designed to build your confidence for the exam.

Based on content from Citizen Developer

Whether you’re a developer, consultant, or architect involved with Power Platform or Microsoft 365 Copilot, this pathway offers insights into the exam’s core objectives, from agent orchestration fundamentals to enterprise data sources and agent evaluations.

Key Study Areas

• Agent Orchestration: Understanding how multiple specialized agents can collaborate effectively in complex scenarios. Agent orchestration isn’t merely about inter-agent communications; it includes how an agent acts as an orchestrator, deciding when to invoke various tools and knowledge resources.
• Knowledge and Enterprise Data Sources: You’ll need to understand how business context can ground your agents, helping them generate appropriate responses based on enterprise data.
• Tools and User Interactions: Familiarize yourself with adaptive cards, custom connectors, and REST APIs, pivotal in extending agents’ capabilities for real-world tasks like notifications and record management.
• Agent Architecture: Knowing the distinction between child agents and connected agents is crucial, as it influences how your AI solutions scale and interact within an organizational framework.
• Evaluation and Lifecycle Management: Application Lifecycle Management (ALM) is a cornerstone, as is understanding agent evaluations and the performance criteria that determine success in real-world applications.

Our starting point examines agent orchestration, a cornerstone of agent ecosystem development. Here, the agent itself acts as an orchestrator, leveraging tools and capabilities to pull knowledge and make decisions, emphasizing how the agent’s role goes beyond simple command execution to involve intelligent decision-making.

When constructing agents, the interplay between knowledge and tools is vital. Where knowledge empowers your agent to answer questions, tools enable it to perform actions like updating records. Grasping this distinction is critical—especially for exam scenarios where nuanced understanding can provide the answer.

Finally, let’s not overlook the importance of integrating agent evaluations into your study. While these evaluations seemed understated in the Microsoft Learn modules, the exam showed them to be significant, making up a considerable portion of the exam content.

Ultimately, if you aim to master the exam, understanding the modular composition of AI agents and their orchestration through tools and knowledge is paramount. Though the AB-620 is challenging, familiarizing yourself with the exam’s landscape can make it a platform for your aspirations in AI-driven solutions.

For additional resources, download the AB-620 Study Guide Slides and explore more from Microsoft’s training courses here.

If you have questions or require more personalized guidance, feel free to arrange a one-on-one session via Calendly. For continued insights and guidance, connect with me on LinkedIn, and visit Citizen Developer to dive deeper into this realm.

Good luck as you embark on your certification journey!

Explore new education plugins for ChatGPT Work and Codex that help K–12 teachers, college educators, and students learn, teach, research, and build.

Over the past two months, we’ve been previewing a new capability within Copilot Studio that lets you build agents capable of taking on more complex business processes. Today, we’re excited to share that this capability is now generally available for production use and to introduce its new name: the GitHub Copilot harness in Copilot Studio. 

 

Meet the GitHub Copilot harness  

The GitHub Copilot harness gives Copilot Studio the coding and reasoning capabilities behind our most advanced agent experiences (like Copilot Cowork and the GitHub Copilot coding agent). This means your agents can now handle work that used to be out of reach, including processes that have many steps, many sources, and ambiguous decision points. 

The new harness is built for complex, long-horizon work using the latest frontier reasoning models like Opus 5, GPT-5.6 Sol, and Fable 5. It can plan, reason through dynamic problems, run an agentic loop, use skills, integrate workflows, connect to tools and agents in other platforms, and produce rich, multi-part outputs.  

In our own testing with real world business process evals, Copilot Studio shows significant performance and quality gains when using the GitHub Copilot harness. Improvements in multi-tool use, file analysis, code analysis, and knowledge quality mean these agents are more capable of taking on your most complex business processes. 

 

To help you build these more capable agents, we’re also delivering new experiences for makers building agents with the GitHub Copilot harness. The agent designer is more intuitive for authoring, putting the most important tools right within reach to make authoring faster, while retaining full agent lifecycle management features.  

A gif showing the process of adding a skill to an agent in the agent designer and starting a preview test conversation

Then, the workflow designer gives a visual canvas to understand and edit workflows, including adding agent nodes and running workflow evals. And soon, natural language authoring will let you describe your business goal and assemble the right combination of agents and workflows through a multi-turn conversation. 

Agents running on the GitHub Copilot harness use usage-based billing for all work, regardless of Microsoft 365 Copilot licensing. You pay for your agent’s usage based on the models you choose, the organizational context and tools you add, and runtime used. Certain AI-driven maker experiences, like natural language authoring, evaluations, and testing, will also fall under usage-based billing if building with the GitHub Copilot harness.  

You can learn more about Copilot Studio usage-based billing here. 

Continued Support for Copilot Chat and Standard harnesses 

Adding a new harness does not change the ones you already rely on. Different harnesses can be optimized for different outcomes, so rather than relying on a single one-size-fits-all approach, Copilot Studio now lets you choose the harness that best fits your scenario.  

This means Copilot Studio supports three harnesses today: 

1. The Copilot Chat harness, which uses the same harness as Microsoft 365 Copilot Chat and is ideal for customizing Copilot Chat experiences.
   
   
2. The Standard harness, which most agents built in Copilot Studio use today and is great for conversational agents with rules-based topics.
   
   
3. The new GitHub Copilot harness, which uses the power of GitHub Copilot SDK to automate complex, agentic business processes. 

We will continue to support the Copilot Chat and Standard harnesses in Copilot Studio for both existing agents and authoring of new agents. Microsoft 365 Copilot licensed users continue to benefit from fair use of Copilot Chat or Standard harness agents included in their Microsoft 365 Copilot license. For all other usage, the Copilot Chat and Standard harnesses will continue to be billed using the existing fixed rate card. 

Get started today 

The GitHub Copilot harness is generally available now, and you can start creating these agents directly from the Copilot Studio homepage. We can’t wait to see what you build. Share your thoughts using the feedback control in the top right corner of the product. Your input directly shapes what we build—and how you build—next. 

Try Copilot Studio today!

At a glance Edge for Business security connectors extend your security tools into the browser, so you gain visibility and enforcement where work happens. This update adds new partner integrations, including Cisco Secure Access, Tanium, and Clever.

---

You’ve built a security stack to protect your organization. As work shifts into the browser, a critical question emerges: can those tools protect work at the point where it happens? Good news, they can—when combined with the Edge for Business security connector framework. The framework extends your existing security stack into the browser, bringing greater visibility and policy enforcement into everyday work without adding another control plane or duplicating tools. We’re expanding this ecosystem with new integrations so you can maintain consistent protection wherever work happens.

Now generally available

• Cisco Secure Access is an advanced enterprise browser integration for Edge for Business. As a single integrated connector, it brings not only secure access but also unified data protection and AI guardrails directly into the browser.
• Tanium Connector for Microsoft Edge for Business delivers real-time browser telemetry into Tanium Autonomous IT Platform, helping organizations gain deeper visibility into browser-based threats, close critical security blind spots, improve compliance validation, and strengthen overall endpoint security across the enterprise.
• Clever Classroom MFA TrustPass helps streamline logins in education environments by recognizing trusted devices in Edge for Business—reducing unnecessary MFA prompts while maintaining strong security.

Bringing it all together

Our goal at Edge for Business is to give you a simpler way to secure how work gets done. As our connector ecosystem continues to grow, you have more options than ever to easily integrate the security tools you already use into the browser. Easily Integrate Security Tools with Connectors

Apple says its trade secrets investigation into OpenAI has widened. In a new court filing, Apple claims additional former staff may have retained or accessed confidential information.