Advanced level · Module 2: AI Agents and Tool Use · Lesson 1 of 4
Tool and Function Calling
Let a model request calls to functions you define, and execute them safely.
In this lesson you will
- Define tools with clear schemas
- Execute tool calls and return results to the model
Key concepts
tool schemafunction callingargument validationtool results
The lesson
With tool calling, you describe functions — name, purpose and parameters — and the model can ask to call them, such as search_policies(query) or get_office_hours(location). Your code runs the function and returns the result to the model.
The model only requests; your code decides. Validate arguments before executing, as you would any user input.
Clear tool descriptions and small, focused tools make correct use more likely.
Step by step
- Define a read-only search tool with a schema.
- Let the model call it for a question.
- Validate the arguments and run it.
- Return the result and get the final answer.
Practice activity
Add the policy search tool to the capstone assistant.
Expected result: An assistant that calls the search tool correctly and answers from its results.
Check your understanding
-
Who actually executes a tool call?
Show answer
Your code, after validating the request.
-
Why keep tools small and focused?
Show answer
Clear, simple tools are used correctly more often.
Taught live by an instructor — online or in person. Free course lessons are for self-study and do not lead to a certificate — see registration options.