Chinakeyboardman commented on PR #25871: URL: https://github.com/apache/pulsar/pull/25871#issuecomment-4557060530
Thanks for putting this together — the split structure is really thoughtful and will definitely help both human contributors and AI agents navigate the codebase more effectively. A few small suggestions for future iterations: 1. Code examples (do / don't) would go a long way The conventions in CODING.md are clear, but AI agents tend to follow concrete code patterns much more reliably than prose rules. A few short before/after pairs for the key conventions (e.g. CompletableFuture method must not throw synchronously, prefer record over Pair , flatten nested futures with thenCompose ) would make the guidance much more actionable for tooling. 2. A few places have duplicated information across files For example, the security reporting steps and the "don't rebase after PR is open" rule appear in both AGENTS.md and CONTRIBUTING.md . Keeping each rule in one canonical place with cross-references would reduce maintenance overhead and avoid the same content being loaded into context twice. None of these are blockers — the foundation here is solid. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
