Building Your Own AI-Powered Jira Tools to Boost Team Productivity

The integration of Generative AI (GenAI) into software development has rapidly become standard practice, particularly with tools like GitHub Copilot, Amazon Q Developer, and hundreds of test automation tools available in the market. However, GenAI's applications extend far beyond just code generation, permeating every phase of the Software Development Life Cycle (SDLC), from defining requirements to generating documentation and test cases. With this in mind, I developed Jira Copilot, an AI-powered Jira tool, as a Proof of Concept (POC) to demonstrate how GenAI can significantly boost team productivity when working with Jira, one of the most widely adopted project management tools in the industry.

Jira Copilot serves as an example of how Generative AI can extend its capabilities beyond code generation. While there are several AI-integrated Jira plugins available, data privacy remains a significant concern for many businesses. My goal with Jira Copilot is to show organizations how easily they can build their own AI-powered Jira tools, ensuring full control over data privacy and security.

As a POC, Jira Copilot offers a set of straightforward features that organizations can explore to assess the potential of GenAI in their workflows:

• Jira Ticket Validation and Recommendations: Jira Copilot can automatically validate whether a Jira ticket — be it a User Story, Bug, or Task — adheres to a predefined format and contains all necessary information. If any details are missing or the ticket can be improved, the AI provides recommendations, ensuring that all tickets meet a high standard of quality.

• Ticket Rewriting: One of the standout features is the ability to rewrite Jira tickets. By leveraging Generative AI, Jira Copilot can reformat a ticket's description to align with the organization's standards and ensure that all required information is included. Users can then choose to update the ticket description with the AI-generated content, streamlining the documentation process.

• Test Case Generation in BDD Format: For user stories, Jira Copilot can generate a list of test cases following the Behavior Driven Development (BDD) methodology using the Gherkin format. This not only aids in clarity and understanding but also facilitates easier automation of these test cases later on.

• Language Translation: In today's globalized development environment with offshore or outsourcing teams, language barriers can sometimes impede progress. Jira Copilot offers a translation feature that allows users to translate ticket descriptions into other languages, such as Vietnamese, broadening the tool's usability across multinational teams.

• Custom Prompts for Extended Functionality: To provide even more flexibility, Jira Copilot supports custom prompts. This feature enables users to instruct the AI model to perform additional actions on tickets, beyond the predefined features. This open-ended capability ensures that Jira Copilot can adapt to a wide variety of team needs.

The features outlined above are just the beginning. Jira Copilot is designed to be an adaptable tool that can be extended to fit specific organizational requirements. As teams explore these initial features, they can identify other areas where Generative AI could enhance their workflows, leading to the development of more complex and customized solutions.

Jira Copilot is developed using the Atlassian Forge framework, utilizing NodeJs for the backend and React for the frontend. This integration ensures a seamless experience within the Jira environment. For Generative AI services, Jira Copilot supports both OpenAI and AzureOpenAI APIs, providing flexibility in choosing the AI infrastructure that best suits an organization's needs.

Jira Copilot is open-source and licensed under the GPL-3.0 license. You can access the project on GitHub.