What if you could design, version, and deploy your Copilot Studio agents straight from your terminal - with GitHub Copilot CLI or Claude Code acting as your true development copilot?

That's exactly what the open-source Skills for Copilot Studio (Microsoft) project enables: a plugin for AI coding tools that exposes Copilot Studio as a genuine code target - topics, actions, knowledge, triggers, and variables, all in YAML, with schema validation, templates, and specialized sub-agents.

In this session, we'll cover:
- The plugin's architecture and its four sub-agents (manage, author, test, advisor).
- How to clone, edit, and push a Copilot Studio agent from the terminal using GitHub Copilot CLI, Claude Code, or VS Code.
- The full workflow: YAML authoring → validation → push → automated testing.
- Best practices for treating an agent like a code project: versioning, CI/CD, quality review.
- Live demos on real-world scenarios, from clone to deployment.

Key takeaways:
-Learn what the Skills for Copilot Studio plugin is and how it exposes topics, actions, knowledge, triggers, and variables as YAML code.
- See a real example of the full agent-as-code workflow: clone → author → validate → push → test, driven from the terminal.
- Understand how to apply software engineering practices — versioning, CI/CD, quality review - to Copilot Studio agents.

Key benefits of attending:
- Know your tools: which sub-agent (manage, author, test, advisor) to use at each stage of building a Copilot Studio agent.
- Learn from real demos: what an industrialized, code-first agent workflow looks like from clone to deployment.