You can create customized AI actions using the Custom template. When you enter the desired AI template description, a template is quickly generated based on it, and the necessary prompts are automatically written.
Action Name: Specify a name for the AI action.
Sample Output Data: Defines the format of the sample output the AI will generate.
JSON: Selected if structured data is expected from the AI. When this option is used, the expected JSON structure and fields can be predefined, so the model generates the output according to the specified schema.
Text: Used for free-form (natural language) outputs.
User Prompt: Represents the input to be received from the user. Specifies what the user wants to ask the AI in this field.
Example: You are an AI template creator assistant working in a low-code development environment. When a user provides a brief description of their desired AI use case, your task is to:
Analyze the use case to understand the objective of the AI action.
Automatically generate an appropriate AI template that includes the following:
Task type (e.g., text generation, classification, summarization, translation, etc.)
Detailed prompt structure with variables (placeholders)
Sample output
Input/output format description
Generate the template in a clear, developer-friendly format and prepare it for integration into the Kuika AI Actions system.
System Prompt: This is the background instruction provided to the AI. This field defines how the model should behave (e.g., “You are a legal expert.”). In other words, it determines the AI’s role.
Choose LLM Model: In the AI Actions data source, select from GPT-5, GPT-5 Mini, GPT-5 Nano, GPT-5.1, GPT-5.2, GPT-5.4, GPT-5. 4 Mini, GPT-5.4 Nano, Gemini 2.5 Flash, Gemini 2.5 Flash Image, Gemini 2.5 Flash Lite, Gemini 2.5 Pro, GPT Image 1, and Veo 3.1. The default model is GPT-4.
Parameters:
Receipt_text: Used by AI Action to analyze, interpret, or convert text into structured data.
Gemini File Parameter: Used to provide file-based input. When “Allow file upload” is enabled, an image or file can be uploaded from a device and sent to the model in Binary (Bytea) format. When disabled, only text-based (String) parameters are used.
Unknown: Unknown or undefined data type.
Boolean: True/false value.
Bytea: Binary data (e.g., image, file, etc.).
Byte: Integer value between 0 and 255.
DateTimeOffset: Date and time information.
Decimal: Decimal number.
Double: Larger decimal number.
Guid: Unique identifier (UUID).
Int16: Small integer value.
Int32: Medium-sized integer.
Int64: Large integer.
Single: Small decimal number.
String: Text value.
AI Features:
Temperature: Set the AI’s creativity level (0 = consistent, 1 = random responses).
Maximum Length: Set the length of the content the AI will generate in tokens.
Stop Sequences: Add a word or sequence of words to make the response stop at a specific point.
Top P: Controls the width of the word pool the model uses when generating output. Takes a value between 0 and 1. Lower values ensure the model selects only the most likely and logical words; higher values allow selection from a broader and more varied word pool.
Frequency Penalty: Used to reduce the repetition of words.
Presence Penalty: Encourages the creation of new content.
Reset to Defaults: Restores modified parameters to their original default settings.
Finally, click the CREATE button in the top-right corner to create your AI action.
