Pattern: skills that strip AI writing tells are appearing in several languages, with at least four created in two days

GitHub·low signalXBlueskyLinkedInCopy link

GitHub topic searches for claude-code and agent-skills turn up leter/zh-tech-writing (148 stars, Chinese technical writing), baibanbao/qu-ai-wei (61, which reconciles seven conflicts between three Chinese rule sets), ilien-dev/quiron (44, which re-scores the rewrite after editing) and ormeilu/avoid-ai-writing-russian. All four were created between 09-23 and 09-24. A separate stdlib-only detector, F0Rextasy/aitell, publishes a confusion matrix and fails CI if its accuracy drops. Prose-style rules are now shipped as installable skills per language, and the stronger ones measure their output instead of only listing banned words.