Skip to main content

Microsoft.Extensions.AI–Part III–Tool calling

I'm on a journey discovering what is possible with the Microsoft.Extensions.AI library and you are free to join. Yesterday I looked at how to integrate the library in an ASP.NET Core application. Today I want to dive into a specific feature; tool calling.

This post is part of a blog series. Other posts so far:

What is tool calling?

With tool calling you are providing your LLM with a set of tools (typically .NET methods) that it can call. This allows your LLM to interact with the outside world in a controlled way. In Semantic Kernel these tools were called ‘plugins’ but the concept is the same.

To be 100% correct it is not the LLM itself that is calling these tools but the model can request to invoke a tool with specific arguments (for example a weather tool with the location as a parameter). It is up to the client to invoke the tool and pass the result back to the LLM.

Remark: You maybe wonder, what about MCP? You can see MCP as a kind of standardized way to do tool calling. But I have planned a separate post where I specifically dive into MCP integration with Microsoft.Extensions.AI.

Integrating tool calling in Microsoft.Extensions.AI

Microsoft.Extensions.AI provides 3 building blocks to add tool calling:

  • AIFunction: The .NET method(aka Tool) that can be described to an AI model and invoked.
  • AIFunctionFactory: A factory class that helps you create AIFunction instances based on .NET methods.
  • FunctionInvokingChatClient: A wrapper for IChatClient that adds automatic function-invocation capabilities.

Let’s put these building blocks in action.

We take our application from last post and continue there:

  • Let us start by creating our .NET method that will be exposed as a tool to our LLM:
  • Now we take our original IChatClient and wrap it in a FunctionInvokingChatClient:
    • Remark: Notice that I’m using Ollama in this example as I couldn’t get Tool calling working when using AI Foundry Local.
  • The last step required is to construct a ChatOptions object and pass our tools:
    • Remark: notice that we provided extra information about the tool when creating the AIFunction.
  • Don’t forget to pass our a ChatOptions object  when calling the GetStreamingResponseAsync method:

Remark: Be aware that not every model supports tool calling. You can try different models to find one that works or use an OpenAI model.

More information

Semantic Kernel – Auto function calling

Labels:

Popular posts from this blog

Podman– Command execution failed with exit code 125

After updating WSL on one of the developer machines, Podman failed to work. When we took a look through Podman Desktop, we noticed that Podman had stopped running and returned the following error message: Error: Command execution failed with exit code 125 Here are the steps we tried to fix the issue: We started by running podman info to get some extra details on what could be wrong: >podman info OS: windows/amd64 provider: wsl version: 5.3.1 Cannot connect to Podman. Please verify your connection to the Linux system using `podman system connection list`, or try `podman machine init` and `podman machine start` to manage a new Linux VM Error: unable to connect to Podman socket: failed to connect: dial tcp 127.0.0.1:2655: connectex: No connection could be made because the target machine actively refused it. That makes sense as the podman VM was not running. Let’s check the VM: >podman machine list NAME         ...
Read more

Cache stampede: when our cache turned against us

While investigating some performance issues, we ran into an ASP.NET Core API that cached a fairly expensive aggregation query for 60 seconds. Under normal load, that was fine: one request rebuilds the cache, everyone else reads from it. Under peak load, dozens of requests would arrive in that same expiry window, all see a cache miss, and all fire the same expensive query in parallel. The database didn't like that. That was the moment when our caching layer stopped helping and started hurting. A burst of requests comes in at the same time, all miss the cache, and all go hammer the database or the downstream API at once. That's a cache stampede . The cache was supposed to protect our backend, and for a few hundred milliseconds it did the opposite. Why this happens IMemoryCache.GetOrCreate (and its async sibling) looks like it protects you, but it doesn't add any locking on its own. Look at the naive version: public async Task<Report> GetReportAsync(string key) ...
Read more

A complex system designed from scratch never works

A few years ago, I worked as an architect on a big mainframe rewrite. I still count it as one of my failures. Not because the technology was wrong, but because I couldn't convince the management team to simplify the approach. Years later, the organization is still struggling to get the new system up and running. I left the project at the time, because I couldn't put my name behind an approach that would take very long and cost a lot of money without a working system to show for it along the way. Gall’s Law That memory keeps coming back to me, because it's a textbook case of Gall's Law playing out in real life. Gall's Law , from John Gall's Systemantics , states it plainly: A complex system that works is invariably found to have evolved from a simple system that worked. A complex system designed from scratch never works, and it cannot be patched to make it work. You have to start over with a simple system that works. What does that mean in practice,...
Read more