From e7b79dd55f8eaebb4138e0e70183e695d695c574 Mon Sep 17 00:00:00 2001 From: Shahak Shama Date: Tue, 10 Mar 2026 17:13:21 +0200 Subject: [PATCH] improve ai section --- style.md | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) diff --git a/style.md b/style.md index 6914b9a941f..b304db86666 100644 --- a/style.md +++ b/style.md @@ -56,13 +56,16 @@ For single-file modules, they should also start with a `foo_` prefix ## AI-Generated Code Guidelines +Code generated by AI is still your code: You should stand behind every line of code you submit for PR. + Take extra care when reviewing ai-generated code for these issues: -- Using deprecated/dead crates (!): for example `lazy_static` often appears in ai-code (see [lazy static section](#lazy_static-crate)). +- Using deprecated/dead crates (!): for example `lazy_static` often appears in ai-code (see [Unmaintained dependencies](#unmaintained-dependencies)). - Using "toy" crates (low github-starred, pet‑project crates not intended for production): just because a crate out there can fulfil a given set of requirements doesn't mean it's production-ready, see also [External Dependencies](#external-dependencies) section. - Excessive code-comment bloat: see [Textual Content Quality](#textual-content-quality) section. - Hygiene issues: leaks and races are common, since chatbots rarely take a step back and analyze the surrounding conditions in which a code is being run; always analyze flows affected by your change as the AI will not do so. - Excessive allocations/clones. +- Resolving merge conflicts: AI tends sometimes to just pick our changes without thinking about it. ## Testing