Junie And Rubmine
I have to say that Junie Pro isn't worth it and Junie Ultimate is very worth it. The disparity between allowance is huge. Pro barely lasted 3 days, Ultimate I had only used about a third of my allowance in a month.
Anyways,
My question is when Junie does a report on my codebase, what code does it look at. I often run a report, make all the changes, update the report and it doesn't notice any changes I have made.
It also seems to know historic data outside of the current codebase and conversation. ie it showed me how code had evolved with before and after examples. Does it go back through git commits?
It's a bit confusing. TIA
Please sign in to leave a comment.
• Break large tasks into smaller ones: Smaller and concrete tasks normally take fewer tokens because Junie requires less context exploration, less requests to LLMs.
• Use follow-ups wisely : Follow-ups and improvements with Junie may take more tokens. To minimize this, be sure to create clear instructions for Junie from the beginning and add them to the guidelines (https://www.jetbrains.com/help/junie/customize-guidelines.html).
• Use Junie and AI Assistant in the same project but for different tasks : We designed Askmode in Junie to collaborate better with the agent, e.g. brainstorm new features or ask for potential solutions. Agent consumes more quota than AI Assistant, so if you have a simple question related to your project which doesn't require working with the agent, use AI Assistant instead.
• Keep your guidelines file concise: If you use a guidelines file (`.junie/guidelines.md`), keep it short and clean to reduce quota usage. While 50 lines vs 100 lines won't make much difference, 100 vs 1000 will.
Junie is designed to autonomously explore the entire project and handle complex tasks. As I understand it does not see specific changes in your project. Could you check if they are not part of .aiignore. It would help if you can share your sample prompt so we can reproduce the issue and file a bug if necessary.
Junie may use local history (the IDE’s internal record of file changes) rather than git status to detect changes. Which may answer how the evolution of code can be seen.