Project Copilot

An AI assistant for Agile Software Project Management, utilizing Project Manager expertise to automate prompt engineering, and assist in creating design documents, epics, user stories, and sub-tasks.

Comment

Project Copilot: AI-Powered Jira and Confluence Assistant

Inspiration

The future of programming lies in the natural language. We believe that user stories and design documents are the perfect mediums to bring this future to life. Our inspiration for Project Copilot comes from the idea that well-planned and well-designed documents can minimize deviations and rework, and accelerate code generation. By using these documents as inputs for AI assistants, we can automate code generation and make the software development process more efficient.

What it does

Project Copilot is an AI assistant that helps you write better epics, user stories, and sub-tasks on Atlassian Jira. During this Hackathon, we've added new features to our Jira App, including the ability to create a semantic description of a pull request or a peer review. We've also added support to compare the semantic description of the pull request or peer review with the user story, allowing for what we call a "semantic review". This is a validation in natural language of the project definition, requirements, planning, and implementation.

We've also integrated Project Copilot with Atlassian Confluence. Now, linked Design Documents from Confluence can be used as a context to write user stories. We've also developed a client app for Confluence to assist in writing and completing software design documents.

How we built it

We built Project Copilot using advanced AI and machine learning techniques. The AI assistant uses related and liked user stories and linked design documents from Atlassian Confluence to generate high-quality content. We've used semantic search algorithms to find related epics, user stories, and subtasks, and to provide a better context for the operations.

Challenges we ran into

Integrating AI with Jira and Confluence was a significant challenge. We had to ensure that the AI assistant could understand and generate content that was contextually relevant and accurate. We also had to ensure that the semantic search algorithms were efficient and effective.

Accomplishments that we're proud of

We're proud of the new features we've added during this Hackathon. The ability to create a semantic description of a pull request or a peer review, and the support for semantic reviews are significant advancements for Project Copilot. We're also proud of the integration with Confluence and the development of the client app for Confluence.

What we learned

We learned a lot about integrating AI with Jira and Confluence. We also learned about the importance of well-planned and well-designed documents in the software development process. We've gained valuable insights into how AI can be used to automate and improve this process.

What's next for Project Copilot

We plan to continue improving Project Copilot. We want to add more features and make the AI assistant even more efficient and effective. We also plan to explore more ways in which we can integrate AI with Jira and Confluence to automate and improve the software development process.

Built With

Share this project:

Updates

posted an update —

Exciting news for all Atlassian users! Our latest blog post provides a comparative analysis of three AI assistants: Chat GPT, Atlassian Intelligence, and Project Copilot, our AI project for the Atlassian platform.

Our findings suggest that while all three AI assistants have their unique strengths, Project Copilot stands out for its comprehensive approach. It not only helps in drafting user stories but also generates acceptance criteria, technical references, and even provides a semantic review of the code. This holistic approach can significantly enhance the quality of software projects and reduce deviations and rework.

We've found that Project Copilot's ability to provide a comprehensive user story, acceptance criteria, and technical reference without requiring a prompt is a game-changer. Moreover, its unique feature of providing a semantic review of the code makes it a powerful tool for software project management on the Atlassian platform.

Check out our full blog post for a detailed comparative analysis and learn how Project Copilot can revolutionize your project management experience on Atlassian. Comparative Analysis of Chat GPT, Atlassian Intelligence, and Project Copilot

Log in or sign up for Devpost to join the conversation.

posted an update —

Hey Everyone,

Exciting developments have been taking place here at Project Copilot! For the Atlassian Hackathon, we've redefined our search capabilities by replacing our internal search with a more powerful non-conventional semantic search engine. This allows for more nuanced and detailed inquiries, permitting us to utilize Confluence documents more effectively to augment and improve our epics, user stories, and sub-tasks.

A critical update is the addition of a new section to our user stories—the semantic validation section. This enables an in-depth comparison between the code generated to implement the user story and the user story itself, making it easier to detect and rectify deviations, thereby enhancing the accuracy of our product output.

Big news for the week ahead! Following the culmination of the hackathon, we are thrilled to launch the latest version of our app. This version is equipped with integrations to Confluence linked documents, representing our commitment to continuous optimization and synergistic collaboration. Furthermore, we are expanding our horizons by launching our first app designed specifically to assist content creation in Confluence.

These milestones reflect our team's dedication to utilize cutting-edge technology and innovative solutions to deliver user-centric offerings. Stay tuned for further updates.

Best,

Matias Molinas Co-Founder and CTO of Project Copilot

Log in or sign up for Devpost to join the conversation.