agent-development

Guide building Microsoft platform AI agents with scaffolding and deployment plans.

Updated Mar 13, 2026
One-click install
npx skills add https://github.com/bmarcurella/claude-marketplace --skill agent-development-bmarcurella
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: agent-development
Source: https://github.com/bmarcurella/claude-marketplace/tree/main/plugins/ms-copilot-platform/skills/agent-development
Command: npx skills add https://github.com/bmarcurella/claude-marketplace --skill agent-development-bmarcurella

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill provides a structured, conversational approach to building AI agents on the Microsoft platform, guiding users from understanding requirements to deploying a functional agent.

Core Features & Use Cases

  • Guided Development: Follows a "Listen, Infer, Recommend, Build" workflow.
  • Path Selection: Helps choose between zero-code, low-code, and pro-code agent types (SharePoint Agent, Agent Builder, Copilot Studio, Declarative Agents, Custom Engine Agents).
  • Scaffolding & Planning: Generates detailed agent plans and leverages the Agents Toolkit for scaffolding.
  • Use Case: A developer needs to build a custom AI agent to manage customer support tickets. This Skill guides them through selecting the right agent type, framework (e.g., Semantic Kernel), language (Python), and deployment strategy, providing a clear plan and execution steps.

Quick Start

Use the agent-development skill to start building a new AI agent for the Microsoft platform.

Frequently Asked Questions about agent-development

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I build a custom AI agent on the Microsoft platform?

Microsoft AI agent development involves selecting an agent type, choosing a framework like Semantic Kernel, scaffolding with the Agents Toolkit, and executing a deployment plan. This Skill provides conversational guidance through each phase.

What is the difference between declarative agents and custom engine agents in Copilot?

Declarative agents use configuration-driven behaviors, while custom engine agents require pro-code development with frameworks like LangChain or Semantic Kernel. This Skill helps you choose the right path based on your requirements.

Can I use Semantic Kernel with Python to develop a Microsoft AI agent?

Yes, Semantic Kernel supports Python for building custom engine agents on the Microsoft platform. This Skill guides you through framework selection, scaffolding, and execution steps for Python-based agent development.

What's the best way to choose between zero-code and pro-code agent types for my use case?

Choosing between zero-code, low-code, and pro-code agent types depends on your customization needs and development resources. This Skill follows a Listen, Infer, Recommend workflow to match your requirements with the right agent path.

Does the Microsoft Agents Toolkit support scaffolding for declarative agents?

Yes, the Microsoft Agents Toolkit supports scaffolding for declarative agents and custom engine agents. This Skill leverages the toolkit to generate detailed plans and provide execution guidance for your agent project.

When should I not use Copilot Studio for building an AI agent?

You should avoid Copilot Studio when your use case requires deep custom logic or specialized framework integration that exceeds low-code capabilities. This Skill helps determine when a pro-code custom engine agent is more appropriate.