Providing Context to the AI Test Case Assistant 1.0
Before starting the test case generation process, the Tracker Administrator can provide additional details to the AI Test Case Assistant, such as a detailed overview of the current tracker and some guidelines for creating the test case. The AI Test Case Assistant uses this information as a context for generating accurate content for the test case.
 |
The Set AI Context option is available only to the Tracker Administrators.
|
To set the AI context:
1. On the Codebeamer homepage, select a tracker under the Trackers tab and open it in the Document View or Document Edit View.
2. Select an item and click Set AI Context on the AI Test Case Assistant start page. The Set AI Context dialog opens.
3. In the Tracker Overview field enter a comprehensive overview for the current tracker. The AI Test Case Assistant uses this overview as a context when generating the test case.
4. In the Test Case Guidelines field, enter guidelines for generating the test case.
5. Click Save to save the AI context.
Guidance on Populating Tracker Overview and Test Case Guidelines Fields
To ensure optimal performance and accuracy, keep your Test Case Guidelines and Tracker Overview inputs short and focused. Clear, well-structured descriptions help the AI quickly understand the context and create more accurate, relevant test cases. Avoid lengthy or overly detailed inputs — clarity and conciseness are key. For instance, when configuring the AI context for a requirements tracker, consider the following guidelines:
• The text in the Tracker Overview field should describe the content and purpose of the requirement tracker, including:
◦ The product or system to which the requirements are related.
◦ The type of specifications stored such as functional, interface, performance.
◦ Any relevant domain or business context.
By sharing this information, you enable the AI to better understand the scope and focus of the requirements, resulting in more tailored and meaningful test case generation.
• The content in the Test Case Guidelines field should describe your standard operating procedures for testing within the specific requirement catalog and level. The guidelines help the AI understand your unique process expectations, such as:
◦ Preferred test structure or format
◦ Coverage expectations
◦ Specific validation rules or constraints
◦ Domain-specific terminology or workflows
Providing this context allows the assistant to tailor its output to better align with your organization’s testing standards.
Key Considerations
• The Tracker Overview and Test Case Guidelines fields accept only plain text content.
• For accurate generation of test cases, the preferred input languages for the Tracker Overview and Test Case Guidelines fields are English and German.
• The generated test case retains the language of the original requirement, even when the provided test case guidelines are in a different language.
• You can copy and paste content from other documents into the Tracker Overview and Test Case Guidelines fields.
• The AI Test Case Assistant generates the test case content using an LLM. If the administrator provides additional context using the Set AI Context option, that context is appended to the input passed to the LLM.
• When deploying a project or inheriting a tracker, the Tracker Overview and the Test Case Guidelines are not carried over automatically. If these elements are essential for your testing process, ensure to manually review and reapply them after deployment or inheritance.