Overview
We introduced the ChatGPT assistant SpiraApp in Spira v7.10 that allows users to automate the generation of standard downstream artifacts from user stories, features, epics, and business/system requirements. It allows users to quickly generate a set of standard test cases, tasks, risks, and BDD scenarios that can then be refined and improved as needed. We were pleasantly surprised by the response from our customers, everyone was excited about the functionality and the potential time savings it could generate.
However for a lot of our customers in more regulated industries, there was concern with sending their data to the public OpenAI API that is needed to generate the artifacts. Although OpenAI's terms of service state that API queries are not used to train the underlying Large Language Model (LLM), many of our customers were interesting in having a more private solution.
So we have good news! The latest release of Spira includes a new Azure OpenAI version of the plugin that lets you connect to your private LLMs hosted inside Microsoft Azure, accessible via. the same REST API as the public ChatGPT version.
How Does It Work?
The new Azure OpenAI SpiraApp is available inside Spira as a separate SpiraApp (from the existing ChatGPT one). Simply Activate the new plugin:
Then you can enter in the settings to enable Spira to connect to your instance of Azure OpenAI. The settings are different to those in the regular ChatGPT version:
Since each instance of the OpenAI GPT models is hosted privately inside a customer's Azure environment, you need to enter the URL of your own instance of Azure OpenAI. In addition, you will need to enter in the API Key and API Version that your environment supports. Once that's done, you can then use the plugin in the same was as the ChatGPT version.
You can choose a specific requirement (including Epics, User Stories, and Use Cases) and then generate the test cases tasks, BDD scenarios and risks associated with them. For example:
What's Planned in The Future?
We are planning on expanding the AI functionality available in Spira to include such use cases as: generating requirements from a product description, generating more detailed test cases and scenarios based on the defined types in the system, and introducing an AI assistant that helps you perform everyday tasks more quickly. We also plan on giving you more flexibility in the choice of LLMs being used, with plans to allow access to models such as Mistral and Claude via. AWS Bedrock.
How Do I Get the New Version?
Download customers can get the latest version right away by going to the secure Customer Area of our website. Cloud customers will have already been updated during the April maintenance window. If you have any questions about the new version, please contact support@inflectra.com. The full release notes can be found on SpiraDocs.