LLM Tool
Overview
The LLM Tool integrates Large Language Models (LLMs) into Dataworkz Agents, enabling them to handle a wide variety of tasks such as generating content, analyzing data, extracting information, or providing automated customer support.
The tool is highly configurable through prompts. By combining system prompts (to define overall behavior and context) and user prompts (to define specific tasks), you can tailor how the LLM responds to different scenarios. This flexibility ensures the Agent produces consistent, high-quality results across use cases.
Prerequisites
Before creating an LLM Tool, ensure that one or more LLMs have been configured on the Dataworkz platform using the LLM Configurator on the Home screen.
Set up
Creating an LLM Tool
To create an LLM Tool, go to Agents > Create Tool and select LLM from the tool options.
About Tool
Start by defining the basics:
Tool Name: Provide a descriptive name.
Description: Explain what the tool is designed to do.
Example: “This tool summarizes long-form text into concise bullet points.”
User Prompt: Define the actual queries or tasks that users can issue.
Example: “Plan a 3-day trip to ${cityname}.”
Prompts can include variables (e.g., ${cityname} or ${input_url}) that automatically become input parameters in the next step.
Input Parameters: Any variables defined in prompts appear here as parameters. Update their type and description to make the tool more intuitive.
Output: The tool generates responses based on the defined prompts. The output format is predetermined but can be described here to clarify what the user will receive.
Validation
After setup, test the LLM Tool by entering values for the input parameters and running the Test Tool. Verify that the system and user prompts work together to produce the desired output.
Last updated