Principle:Cohere ai Cohere python Tool Schema Definition
| Property | Value |
|---|---|
| Type | Principle |
| Domain | Tool Use, Function Calling, Agentic Patterns |
| Source | Cohere Python SDK |
| Last Updated | 2026-02-15 |
| Implemented By | Implementation:Cohere_ai_Cohere_python_ToolV2_Definition |
Overview
A schema specification pattern for declaring available tool functions that language models can invoke during chat interactions.
Description
Tool Schema Definition is the process of describing external functions in a structured format that the language model understands. Each tool is defined with a name, description, and JSON Schema parameters. The model uses these schemas to decide when and how to call tools. This enables function calling / agentic behavior where the model can interact with external systems (APIs, databases, calculators) by generating structured tool call requests.
Usage
Define tool schemas before sending chat requests where you want the model to have access to external functions. Each tool needs a clear description and well-defined parameter schema for the model to use it effectively.
- Define each tool with a unique name that describes the function
- Provide a description that explains what the tool does and when to use it
- Specify parameters using JSON Schema to describe the expected input structure
- Pass the list of tool schemas in the tools parameter of the chat request
Theoretical Basis
Tool use implements the ReAct (Reasoning + Acting) paradigm where language models alternate between reasoning about what to do and taking actions via tool calls. JSON Schema provides a standardized vocabulary for describing data structures, enabling the model to generate valid function arguments.
- The model inspects tool schemas at inference time to understand available capabilities
- Well-defined schemas reduce hallucinated or malformed tool calls
- The JSON Schema standard ensures interoperability with existing validation tooling