Inspiration
During code reviews, teams often waste time figuring out where to focus — especially on bugs, security issues, or high-priority tasks. I wanted to build a tool that instantly highlights risks and gives actionable checklists right inside Jira issues, helping reviewers go faster and catch problems early. The Williams Racing Edition theme inspired me to create something that "accelerates" software delivery without compromising quality.
What it does
AI Code Review Insights adds a smart panel to any Jira issue. It automatically analyzes the issue (type, priority, summary, description, comments) and provides:
Overall risk level (Low/Medium/High) and estimated review time Priority actions to take first Tailored checklists for code review, security, performance, and testing Instant loading with a refresh button and an "AI Enhance" option (ready for future Rovo AI integration)
It helps developers and reviewers quickly understand what to focus on, making code reviews safer and more efficient — all without leaving Jira.
How we built it
Built entirely with Atlassian Forge:
Frontend: React app using @forge/bridge for seamless Jira integration Backend: Node.js resolver functions with smart rule-based analysis Storage: Forge storage for caching results Secure and fast: No external APIs, fully "Runs on Atlassian" Deployed and tested on a live Jira site using the latest Forge CLI and Node.js 22 runtime.
Challenges we ran into
Getting the Forge CLI updated and resolving global installation issues on Windows Parsing Jira's rich text description fields correctly Balancing instant performance with deep insights (decided on fast rule-based analysis with future-proofing for Rovo AI) Working solo under a tight hackathon deadline while learning advanced Forge features
Accomplishments that we're proud of
Delivered a fully functional, polished app that works instantly on any Jira issue Smart detection of bugs, security keywords, and priority levels for tailored advice Clean, modern UI with risk badges, icons, and clear sections — feels like a native Atlassian feature Completed and deployed a working Forge app from scratch in just a few days
What we learned
Deep dive into Forge resolvers, storage, and Jira API field structures How to build responsive, theme-aware UIs in constrained iframe environments Importance of fast initial load vs. heavy AI processing in productivity tools The power of rule-based systems when paired with good UX — sometimes "smart" is better than "full AI"
What's next for AI Code Review Insights
Integrate with Atlassian Rovo AI (using Forge LLM) for deeper, contextual analysis via the "AI Enhance" button Add support for linked pull requests (Bitbucket/GitHub) to analyze actual code changes Create a Rovo Agent version so users can @mention "Code Review Buddy" in chats Expand checklists based on project type (frontend, backend, mobile, etc.) Open source parts of it to help the Atlassian developer community
Log in or sign up for Devpost to join the conversation.