{"skill":{"slug":"essay-humanizer","displayName":"Essay Humanizer","summary":"Rewrite AI-drafted essays into more human-like academic prose. Fine-tuned LoRA over Qwen3-8B guided by 24 Wikipedia-style AI-writing pattern weights plus MDD...","tags":{"corpus-linguistics":"1.0.0","education":"1.0.0","english":"1.0.0","essay":"1.0.0","humanize":"1.0.0","latest":"1.0.2","style-editing":"1.0.0","writing":"1.0.0"},"stats":{"comments":0,"downloads":174,"installsAllTime":0,"installsCurrent":0,"stars":0,"versions":3},"createdAt":1773980540298,"updatedAt":1774012427918},"latestVersion":{"version":"1.0.2","createdAt":1774004514679,"changelog":"- Major update: Skill now includes a fine-tuned LoRA adapter and inference script for local model use (Apple Silicon only).\n- Adds LoRA adapter weights, pattern references, training documentation, and inference scripts.\n- Local inference supported via `Qwen3-8B-MLX-4bit`, with setup instructions and automated adapter decoding.\n- Optional FastAPI server included for HTTP-based remote use or tool integration.\n- Documentation and references extended; several previous documentation-only files replaced with executable scripts and model assets.\n- Still outputs plain text only, with strict prose, math, and citation requirements unchanged.","license":"MIT-0"},"metadata":null,"owner":{"handle":"kevin0818-lxd","userId":"s179hhzw0zteyz1ae4wjft1kjh83e23s","displayName":"kevin0818-lxd","image":"https://avatars.githubusercontent.com/u/75295236?v=4"},"moderation":null}