Building AI Agents with Semantic Kernel, MCP Servers, and Python
Microsoft Developer presents an in-depth look at building AI agents using Semantic Kernel, MCP Servers, and Python, with demonstrations led by Evan Mattson and resources for hands-on experimentation.
Building AI Agents with Semantic Kernel, MCP Servers, and Python
Hosted by Microsoft Developer, this episode continues the exploration of the Semantic Kernel AI Agents Toolbox. Evan Mattson demonstrates how to build intelligent agents by integrating Semantic Kernel with MCP Servers and Python.
Key Highlights
- Invoking MCP Tools: Learn how MCP (Microsoft Cloud Platform) tools are leveraged during the construction of AI agents with Semantic Kernel.
- Using Different MCP Servers: The episode covers both the GitHub MCP Server and the Microsoft Learn MCP Server, showing their distinct roles in agent workflows.
- Azure AI Foundry Integration: Find out how Azure AI Foundry can be integrated into these workflows for advanced AI capabilities.
- Hands-on Demos:
- Setting up MCP plugins within Semantic Kernel
- Executing sample AI agent workflows using Python
- Real-world examples to apply in your own projects
Resources
- Semantic Kernel Project (GitHub)
- Official Documentation and Updates
- Microsoft Learn: Add MCP Plugins
- MCP Samples: semantic-kernel/python/samples/concepts/mcp
- Getting Started with Agents: Azure AI Agent Samples
- SemanticClip Repo
Connect with Hosts
Subscribe for more open-source AI content and submit your OSS project for potential feature at Open at Microsoft.
Additional Info
- New episodes on Tuesdays
- Explore the Open at Microsoft Playlist
- Part 1 available: Building AI Agent Workflows with Semantic Kernel
This episode is ideal for developers looking for actionable guidance on AI agent construction within Microsoft's ecosystem, offering practical demos, sample code, and useful community links.