Tools and Agents

Premium Content

The lesson introduces how LLMs use tools to access real-time information, take actions, and ground responses in facts. It explains the harness pattern where the LLM only outputs text and the harness runs tools, covers agent safety and sandboxing, and surveys the four built-in tools in the Responses API.

Previous Next

Previous Module The Principles of Responsible AI

This Module Develop a Generative AI Chat App

Develop a Generative AI Chat App with Microsoft Foundry

Choosing an Endpoint and SDK

Comparing ChatCompletions and Responses API

Comparing Per-Turn Code Patterns

Tools and Agents

Using the Code Interpreter Tool

Using Web Search and File Search Tools

Using the Function Tool

Next Module Optimizing Generative AI Chat Apps