Add class-based skill to an AI agent in .NET Agent Framework

Authorised Territory demonstrates how to add a class-based skill to an AI agent built with the .NET Agent Framework, using a locally running LLM via Ollama and building on a previously created starter solution.

Full summary based on transcript

What the video covers

• How to add a skill implemented as a class to an AI agent created with the .NET Agent Framework.
• How the agent is configured to use a locally running LLM through Ollama.
• How this video builds on a starter solution created in an earlier video: https://youtu.be/gfTbmQ28zcg

Prerequisites / starting point

• A starting solution created in the previous video (linked above).
• Ollama installed and running locally with an LLM available for the agent to use.

Adding a class-based skill to the agent

• The presenter shows how to implement a skill as a C# class and register/attach it to the agent.
• The goal is for the agent to be able to call into that class as part of its responses (i.e., use the skill as a callable capability).

Running the agent with the local model

• The agent is run against the local LLM provided by Ollama.
• The presenter demonstrates that the agent can use the newly added class-based skill during execution.