AI Agents and Utilities - Overview

Introduction

Exciting tools that drastically shorten the time spent wrangling data, building supply chain models for Cosmic Frog, and analyzing outputs of these models are now available on the Optilogic platform.

Collectively, the Optilogic agentic AI tools are called Ada. This is after Ada Lovelace, widely regarded as the world’s first computer programmer and one of the earliest visionaries to recognize the potential of computational systems beyond pure calculation.

This documentation briefly explains how to access these AI Agents and Utilities, lists the available tools with a short description of each, and provides links to detailed documentation for several of these tools.

Helpful Resources

Before we dive into how to access the AI Agents & Utilities, here are a few links you may find helpful:

• Basic knowledge of DataStar and chatting with Ada is recommended. Start with:
  
  • The introductory video in the "Getting Started with DataStar: Application Overview" Help Center article.‍
  • The Getting Started with Ada & Agentic AI Help Center article.
• Please see the "AI Agents: Architecture and Components" Help Center article If you are interested in understanding how the Optilogic AI Agents work at a detailed level.
• These are the direct links to detailed documentation for several of the tools (they are also provided in the tools list in the last section):‍
  
  • Modeler AI Agent‍
  • Model Outputs Insights AI Agent‍
  • Data Cleansing AI Agent‍
  • Data Profiler AI Agent‍
  • Full Truckload Costing Utility‍
  • Less-Than-Truckload Costing Utility

Accessing Agents

Four of the available agents can be accessed by chatting with Ada and all of them can be accessed by using Run AI Agent tasks in DataStar.

Chat with Ada

When chatting with Ada on the next generation Optilogic platform, users can select the agent they want to use for their prompt:

Please note that:

• The Modeler agent can use the Model Output Insights agent as a tool. Therefore, you can access the functionality of the Model Output Insights agent through the chat UI by selecting the Modeler Agent as the prompt's agent.
• Similarly, the Data Cleanser agent can use the Data Profiler agent as a tool. Therefore, you can access the functionality of the Data Profiler agent through the chat UI by selecting the Data Cleanser Agent as the prompt's agent.

Please refer to the detailed documentation on the individual agents and the getting started with Ada & Agentic AI article to learn more about using these when chatting with Ada.

DataStar

At a high level, the steps in DatsStar are as follows (screenshots follow beneath):

1. Open the DataStar application on the Optilogic platform
2. Create a new project in DataStar, see the "Creating Projects & Data Connections" section in the DataStar Overview documentation.
3. Create a new macro in the project.
4. From the Tasks tab on the right, drag and drop a Run AI Agent task onto the macro canvas.
5. In the configuration of the task, choose the Agent you want to use from the list in the Select Utility section.
6. Configure the Agent inputs in the Configure Utility section.
7. Optionally configure the Run Configuration and Notes sections.
8. Run the macro or just the Run AI Agent task by itself to execute the selected Agent.

Your macro canvas will look similar to the following screenshot after step #4:

After adding a task, its configuration tab is automatically shown on the right-hand side. Give the task a name, and then select the Agent you want to use from the list of available Agents in the Select Utility section. You can also use the Search box to quickly find any Agent that contains certain text in its name or description. Hover over the description of an Agent to see the full description in case it is not entirely visible:

Once an Agent has been selected by clicking on it, the Configure Utility section becomes available. The inputs here will differ based on the Agent/Utility that has been selected. In the next screenshot the Configure Utility section of the Modeler Agent is shown:

Provide the inputs for at least the required parameters, and if desired for any optional ones. Note that hovering over a blue question mark icon will bring up a hover box with a description of the parameter.

Accessing Utilities

Using Utilities works in the same way as using AI Agents, just through the Run Utility task instead of the Run AI Agent task. The following 3 screenshots show 1) a Run Utility task added to a Macro, 2) its Select Utility section, and 3) the Configure Utility section of the Duplicate Macro utility:

Adjust Resource Size

Resource Size for both Run AI Agent and Run Utility tasks can be set in the Run Configuration section, which is indicated as optional. However, for most agents and utilities, the default 3XS Resource Size is not sufficient. It is recommended to update this to XS:

Available Agents & Utilities

The folloing AI Agents and Utilities are currently available. More are being added as they come available. For each a short description is given and for those that have more detailed documentation to go with them, a link to this documentation is included.

AI Agents

• Model Output Insights Agent: Automatically analyzes your Cosmic Frog Model outputs and generates a report. Detailed documentation for this agent can be found here.
• Data Profiler: Run the full data profiler pipeline with statistical analysis, LLM-powered descriptions, quality scoring, alerts, and PK/FK discovery. Detailed documentation for this agent can be found here.‍
• Data Cleansing Agent: Run an AI-powered data cleansing agent with natural language prompts to intelligently analyze and fix data quality issues. Detailed documentation for this agent can be found here.‍
• Modeler Agent: Run the modeler agent with a natural language query to summarize workflow results or answer questions about the current state of the model. Detailed documentation for this agent can be found here.‍
• Air Express Freight Costing: Run the Air Express freight Costing workflow.Date Quality & Standardization: Analyze date format issues and optionally standardize dates to a target format.‍
• Full Truckload Costing: Run the Full Truckload Costing workflow. Detailed documentation for this agent can be found here.‍
• Less-Than-Truckload Costing: Run the Less-Than-Truckload Costing workflow. Detailed documentation for this agent can be found here.

Utilities

• Run Macro: Executes a DataStar macro in a specified project. Connects to the project, retrieves the macro, runs it, and waits for completion. Reports run status and errors if execution fails.
• Duplicate Task: Duplicates a DataStar task within the same macro or to a different macro. Creates a timestamped copy with all configurations preserved. Both target project and target macro must be provided together if specified. If target project and target macro are not specified, the task is copied to the same macro it exists in.
• Convert Data Guru Project into DataStar: Converts Data Guru logic documented in DG Documenter into DataStar. Reach out to support@optilogic.com for access to DG Documenter and for help migrating Data Guru to DataStar.
• Convert Currency: Converts currency values in a sandbox table column using current exchange rates. Creates a new column with converted values. Non-numeric values are saved as NaN. Requires storing a secret with key name 'DATASTAR_EXCHANGE_RATE_API_KEY' and value fetched from https://www.exchangerate-api.com/.
• Run Distance Lookup: Runs distance lookup on a Cosmic Frog model and stores distance records in the model's TransitMatrix table. Supports multiple geo providers (GreatCircle, PCMiler, Bing, Azure, OLRouting), configurable distance units, arc defining tables, and routing preferences.‍
• Swap Connections: Replaces connection references across all tasks in a DataStar project. Updates source/destination connections in Import/Export tasks and connection in RunSQL tasks. Validates connections exist and are the same type. Automatically updates table paths for file-based connections.
• Table Lineage Scan: Analyzes table usage across all tasks in a DataStar project. Categorizes usage as source (task reads from table), destination (task writes to table), or sql (table referenced in SQL query). Provides high-level summary and detailed task breakdown with execution order.‍
• Geocode Model Table: Geocodes a table in a Cosmic Frog model (Suppliers, Facilities, or Customers). Waits for completion and reports results by default. Supports fire-and-forget mode to skip waiting.‍
• Run Demand Model: Executes demand modeling workflow tasks including hierarchy generation, forecasting, and reconciliation. Supports CSV, Parquet, and DataStar adapters for flexible data I/O.
• Solve Model: Solves a Cosmic Frog model by running one or more scenarios. Runs all scenarios by default or a specified list. Supports custom engine, resource size, and infeasibility check options. Waits for completion and reports results.
• Run Integrity Checker: Runs integrity checker against a Cosmic Frog model and writes results to DataStar. Validates the entire model or a single table, then saves table errors, run metadata, and table status to the project's sandbox.
• Duplicate Macro: Duplicates a DataStar macro within the same project or to a different project. Creates a timestamped copy with all tasks and configurations preserved. Creates target project if it doesn't exist.