AI Chat Copilot is a chat pane in Astah. Describe what you want in plain language, and the AI creates, refines and reviews diagrams in your model, while you keep working in Astah. For use cases with prompts and videos, see AI Modeling in Astah.
AI Chat Copilot creates a UML class diagram from requirements, then adds a Reservation class. Edited to shorten waiting time. Results vary depending on the AI service and model you use.
Overview
Supported products
AI Chat Copilot works with these products, version 11 and later:
AI Chat Copilot can create and edit these diagrams in the latest version of each product.
Diagram
Products
Class diagram
Astah Professional, Astah UML
Use case diagram
Astah Professional, Astah UML
Sequence diagram
Astah Professional, Astah UML
Activity diagram
Astah Professional, Astah UML
State machine diagram
Astah Professional, Astah UML
ER diagram
Astah Professional
Requirement diagram
Astah SysML, Astah System Safety
Block definition diagram
Astah SysML, Astah System Safety
Internal block diagram
Astah SysML, Astah System Safety
Parametric diagram
Astah SysML, Astah System Safety
Supported AI services
OpenAI
Anthropic Claude
Google Gemini
OpenAI-compatible services
Azure OpenAI is supported in Astah Professional and Astah UML version 12 and later.
API key required
Astah has no built-in AI. To use AI Chat Copilot, you need your own API key from a supported AI service, and the AI provider bills you for usage. See AI Settings.
Privacy and limitations
Your prompts and model information are sent to the AI service you choose. Change Vision does not receive them. For data and privacy, costs and limitations, see Using AI in Astah.
Get started
Open a project Create a new project or open an existing one.
Open the Chat Copilot pane Choose [View] > [Show/Hide Extra View], then select the Chat Copilot tab.
Set up the AI service Click Settings in the Chat Copilot pane, select the AI service, enter your API key and choose a model. For details, see Settings.
Send a prompt Type your request in User Prompt and click Send Prompt. The response appears in Output, and the elements the AI creates appear in your diagram. To try a prompt right away, see Examples.
Tip: To start over without the previous conversation, click New Chat in the upper-right corner.
Examples
Select a use case to see prompts you can copy and try in your own project.
Open your project in Astah and click the [Chat Copilot] tab.
Copy this prompt, edit the requirements, and send it:
Read the requirements below and create a UML class diagram named
"Library - Domain Model" with the main classes, attributes, operations,
relationships and multiplicities.
Requirements:
- Members can borrow up to 5 books at a time.
- Each book has an ISBN, a title and one or more authors.
- A loan records the borrow date, the due date and the return date.
Review the diagram and ask for changes in the same conversation, for example:
Add a Reservation class so that members can reserve books that are on loan.
This video has been edited: waiting time was shortened and the diagram layout was tidied by hand. Generation time and results vary depending on the AI service and model you use.
Available in Astah Professional and Astah UML.
Review and document a design with AI
Open the diagram in Astah and click the [Chat Copilot] tab.
Ask for a review without changing the model:
Review the class diagram "Library - Domain Model". List missing elements,
unclear relationships and naming inconsistencies, most important first.
Do not change the model yet.
Apply the suggestions you agree with, by number:
Apply suggestions 2, 3 and 4.
Document the design for people who do not read UML:
Add a note to the diagram that summarizes this model in plain language
for team members who do not read UML.
To make sure the AI only reads your model, turn off Edit Mode.
This video has been edited to shorten waiting time. Generation time and results vary depending on the AI service and model you use.
Available in all four products.
Create SysML diagrams with AI
Open your project in Astah SysML or Astah System Safety and click the [Chat Copilot] tab.
Copy your requirements from your specification or spreadsheet, including the header row.
Paste your requirements into this prompt and send it:
Create a SysML requirement diagram from the requirements below.
Use the ID, name and text of each row, and use the Derived from column
for derive relationships. Show all relationships on the diagram.
[Paste your requirements here, including the header row]
Create the system structure:
Create a SysML block definition diagram for the delivery robot with the
blocks DeliveryRobot, Chassis, DriveSystem, BatteryPack and Sensor.
DeliveryRobot is composed of the other blocks.
Trace the requirements to the design:
In the requirement diagram, add satisfy relationships from each block
to the requirements it fulfills.
This video has been edited to shorten waiting time. Generation time and results vary depending on the AI service and model you use.
Click Settings in the Chat Copilot pane, then configure the sections below.
AI Settings
AI Service: the AI service to use: OpenAI, Azure OpenAI, Claude, Gemini or OpenAI-compatible.
Azure OpenAI is available in Astah Professional and Astah UML version 12 and later. Enter your Deployment Name and API Version. Leave API Version empty to use the latest version.
URL: the base URL of the API. You can edit it only for OpenAI-compatible services and other custom endpoints.
Model Name: the AI model to use. After you enter your API key, click [Apply] to load the list of available models, including reasoning models.
Context Window Size: the maximum number of tokens kept as conversation context. Leave it empty to let Astah set it from the model’s limit.
Note: Astah Professional and Astah UML version 12 no longer have a temperature setting. AI Chat Copilot sets it automatically.
General
Prompt History Size: the number of past prompts kept in the prompt history list. The history is saved only on your computer, and responses are not saved. To delete it, click Clear Prompt History.
Edit Mode (on by default): lets the AI add, change and connect elements in your model. Turn it off to let the AI only read and analyze your model.
Debug Mode: writes detailed logs for troubleshooting connection problems. Log file: (User Home)/.astah/professional, uml, sysml or systemsafety/astah-copilot.log
Allow Local File Access: lets the AI read text files on your computer for context. Click [Edit] to choose the folders it can read. Only allow folders that may be shared with the AI service. See Data and privacy.
Proxy
Use System Proxy Settings: connects to the AI service through your system’s proxy settings.
MCP
If you use AI Chat Copilot only inside Astah, you do not need to configure this section. Configure it to connect Astah to external AI clients through the MCP server. The MCP server also uses the AI Settings above (AI service and API key).
Versions and updates
AI Chat Copilot is a plugin that comes with Astah. It includes the MCP server, so updating AI Chat Copilot also updates the MCP server.
Check your version
Choose [Plugin] > [Installed Plugin].
Find Astah AI Chat Copilot in the list and check the Version column.