feature: added agent rules for clean jetbrains kotlin code

[adietish]

provides rules for clean kotlin/jetbrains plugin code when using claude code and cursor

[coderabbitai]

📝 Walkthrough

Walkthrough

Adds a Kotlin and JetBrains plugin standards document, links Cursor to it through a symbolic link, and imports it into CLAUDE.md.

Changes

Kotlin development rules

Layer / File(s)	Summary
Kotlin and plugin rule definitions .claude/rules/jetbrains_kotlin.md, .claude/rules/jetbrains_kotlin.md2	Defines scoped Kotlin coding, JetBrains plugin architecture, error handling, documentation, UI copy, refactoring, and response-format rules.
Editor rule linking .cursor/rules/jetbrains_kotlin.mdc	Replaces the Cursor rule content with a symbolic link to the shared Markdown rule.
Claude rule import CLAUDE.md	Imports the shared Kotlin rule document into Claude configuration.

Estimated code review effort: 2 (Simple) | ~10 minutes

Suggested reviewers: azatsarynnyy, vrubezhny

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Title check	✅ Passed	The title matches the main change: adding JetBrains/Kotlin agent rules for clean code.
Description check	✅ Passed	The description is directly related to the new Kotlin/JetBrains rules for Claude Code and Cursor.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 24.55%. Comparing base (71098f6) to head (9d2b139).
⚠️ Report is 388 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##            main     #339       +/-   ##
==========================================
+ Coverage   0.00%   24.55%   +24.55%     
==========================================
  Files          4      104      +100     
  Lines         26     4414     +4388     
  Branches       0      846      +846     
==========================================
+ Hits           0     1084     +1084     
- Misses        26     3156     +3130     
- Partials       0      174      +174     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[rgrunber]

If this works for you, then go ahead and merge.

Some suggestions for the future :

• There may be certain things under Naming that could be handled more deterministically, using a formatter
• I would consider adding some lint based setup for the Type Safety requirements, if possible
• There should be a setting in IDEA (I know Eclipse has it) that lets you configure the Class Structure. For example, you can sort : Types, Static Fields, Static Initializers, Static Methods, Fields, Initializers, Constructors, Methods .

[adietish]

@rgrunber good hints, appreciate, will include changes for these.
The last item is imho moot. Class structure as displayed in JetBrains (or Eclipse) does not matter here: what I try to achieve is to improve/correct the code that cursor/claude is generating. Do I miss something?

[rgrunber]

Not class structure as it is displayed, but the actual generated code. Various formatters support re-ordering all members in a class based on your specification. Eg. eclipse-jdtls/eclipse.jdt.ls#2169 . Rather than telling an LLM, you can just give it the formatter file and maybe some tool to run it and just tell it "Run this at the end to ensure proper formatting".

[adietish]

@rgrunber ok grocked. But then I'd argue that things should be "fixed" as early as possible in the chain (rahter in the agent than later in the formatter). Of course being a statistical machine agents wont always respect your rule and that's why I'd want a (deterministic) formatter on top.