AI Tool Schema Builder
Build function calling schemas for OpenAI and Anthropic Claude visually. Add parameters, set types, and export JSON.
Features
- ✓Generate function calling schemas for OpenAI and Anthropic formats
- ✓Visual form — add, remove, and configure parameters without writing JSON
- ✓Support for all JSON Schema types: string, number, integer, boolean, array, object
- ✓Toggle required/optional per parameter
- ✓Real-time JSON preview updates as you build
- ✓One-click toggle between OpenAI and Anthropic output formats
How to Use
- 1Choose the output format: OpenAI or Anthropic
- 2Enter the function name and description
- 3Add parameters with name, type, description, and required flag
- 4The JSON schema updates in real time as you fill in fields
- 5Copy the generated schema and use it in your API calls
Examples
Input
Name: get_weather | Params: location (string, required), unit (string, optional)
Output
{"type":"function","function":{"name":"get_weather","parameters":{"type":"object","properties":{"location":{"type":"string"},"unit":{"type":"string"}},"required":["location"]}}}Input
Name: search_docs | Params: query (string, required), limit (integer, optional)
Output
{"name":"search_docs","input_schema":{"type":"object","properties":{"query":{"type":"string"},"limit":{"type":"integer"}},"required":["query"]}}Why Use a Tool Schema Builder?
Function calling (also called tool use) lets LLMs invoke external functions during a conversation. To use this feature, you need to define a JSON schema that describes each function — its name, what it does, and what parameters it accepts. Writing these schemas by hand is tedious and error-prone, especially for functions with many parameters.
OpenAI and Anthropic use slightly different schema formats. OpenAI wraps the function definition inside a {"type": "function", "function": {...}} envelope with a "parameters" key, while Anthropic uses a flatter structure with "input_schema" instead of "parameters". This tool generates the correct format for whichever provider you are targeting.
The visual form eliminates the need to remember JSON Schema syntax. You add parameters by clicking a button, set their type from a dropdown, write a description, and check a box if they are required. The JSON output updates in real time, so you can see exactly what the schema looks like as you build it.
Well-defined tool schemas dramatically improve how reliably LLMs use your functions. Clear parameter descriptions help the model understand what values to pass, and marking parameters as required prevents the model from omitting essential arguments. Investing time in schema quality pays off in fewer failed function calls.
All processing happens in your browser. Your function definitions and parameter details are never sent to any server, making this safe for internal tools, proprietary APIs, and confidential function definitions.