ActionDescriptor class
Describes an action the LLM wants to execute on the UI.
An ActionDescriptor is parsed from the LLM's function-call response. It maps to a registered action in ActionRegistry.
Constructors
-
ActionDescriptor({required String actionName, Map<
String, dynamic> args = const {}}) -
const
-
ActionDescriptor.fromJson(Map<
String, dynamic> json) -
Parse from LLM JSON response.
factory
-
ActionDescriptor.fromToolCall(Map<
String, dynamic> json) -
Parse from OpenAI tool_call format.
factory
Properties
- actionName → String
-
Action name (must match a registered action, e.g. "tap", "enterText").
final
-
args
→ Map<
String, dynamic> -
Arguments for the action (e.g. {"id": "42", "text": "Hello"}).
final
- hashCode → int
-
The hash code for this object.
no setteroverride
- runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
toJson(
) → Map< String, dynamic> -
toString(
) → String -
A string representation of this object.
override
Operators
-
operator ==(
Object other) → bool -
The equality operator.
override