Interface FunctionCallback
- All Known Implementing Classes:
FunctionCallbackWrapper,FunctionCallingHelper.FunctionDefinition,FunctionInvokingFunctionCallback,MethodInvokingFunctionCallback
public interface FunctionCallback
Represents a model function call handler. Implementations are registered with the
Models and called on prompts that trigger the function call.
- Author:
- Christian Tzolov
-
Nested Class Summary
Nested ClassesModifier and TypeInterfaceDescriptionstatic interfaceBuilder for creating aFunctionCallbackinstance.static interfaceFunctioninvoking builder interface.static interfaceMethod invoking builder interface. -
Method Summary
Modifier and TypeMethodDescriptionstatic FunctionCallback.Builderbuilder()Creates a newFunctionCallback.Builderinstance used to build a defaultFunctionCallbackinstance.Called when a model detects and triggers a function call.default Stringcall(String functionInput, ToolContext tooContext) Called when a model detects and triggers a function call.getName()
-
Method Details
-
getName
String getName()- Returns:
- Returns the Function name. Unique within the model.
-
getDescription
String getDescription()- Returns:
- Returns the function description. This description is used by the model do decide if the function should be called or not.
-
getInputTypeSchema
String getInputTypeSchema()- Returns:
- Returns the JSON schema of the function input type.
-
call
Called when a model detects and triggers a function call. The model is responsible to pass the function arguments in the pre-configured JSON schema format.- Parameters:
functionInput- JSON string with the function arguments to be passed to the function. The arguments are defined as JSON schema usually registered with the model.- Returns:
- String containing the function call response.
-
call
Called when a model detects and triggers a function call. The model is responsible to pass the function arguments in the pre-configured JSON schema format. Additionally, the model can pass a context map to the function if available. The context is used to pass additional user provided state in addition to the arguments provided by the AI model.- Parameters:
functionInput- JSON string with the function arguments to be passed to the function. The arguments are defined as JSON schema usually registered with the model. Arguments are provided by the AI model.tooContext- Map with the function context. The context is used to pass additional user provided state in addition to the arguments provided by the AI model.- Returns:
- String containing the function call response.
-
builder
Creates a newFunctionCallback.Builderinstance used to build a defaultFunctionCallbackinstance.- Returns:
- Returns a new
FunctionCallback.Builderinstance.
-