Tool / Function-Schema Builder

What this tool does

Function calling, also called tool use, lets a language model invoke your code. You describe a function with a name, a description, and a parameter schema written in JSON Schema. The model then emits a structured call that your application runs. Every major provider wraps the same JSON Schema in a slightly different envelope. This builder lets you define the function once and export it in all three.

The three formats

• OpenAI (Chat Completions): a function object with name, description, and parameters.
• Anthropic (Claude): a tool object with name, description, and input_schema.
• MCP (Model Context Protocol): a tool object with name, description, and inputSchema in camelCase.

The underlying parameter schema is identical in all three. Only the wrapper changes.

How to use it

1. Enter the function name and a clear description. The description is how the model decides when to call the tool, so make it specific.
2. Add parameters. For each, set a name, a type (string, number, integer, boolean, array, object, or enum), a description, and whether it is required. For an enum, list the allowed values.
3. Copy the format you need from the tabs.

Why descriptions matter

The model reads your names and descriptions, not your code. A vague description leads to the tool being called at the wrong time or with the wrong arguments. Be explicit about when to call the function and what each parameter means.

Related

See how function calling fits into the bigger picture in our guide to the Model Context Protocol, which standardizes how tools are discovered and connected across applications.