NulightJens
Community@NulightJens
NulightJens provides two-pass text editing skills that remove surface-level and structural signs of AI-generated writing from professional content.
Agent Skills by NulightJens
Showing 2 vetted skills indexed across 1 GitHub repositories.
Frequently Asked Questions About NulightJens
FAQPage SchemaWhat tasks can I accomplish using NulightJens's skills?▼
You can remove signs of AI-generated writing from text in two passes: the humanizer skill fixes surface patterns like promotional language, vague attributions, and AI vocabulary, while structural-humanizer removes discourse-level tells such as stated lessons, linear arcs, and shape convergence.
Who should use the humanizer and structural-humanizer skills?▼
Writers, editors, and content professionals producing LinkedIn posts, course lessons, blog posts, essays, newsletters, or emails that must read as human-written. The skills trigger on requests like 'humanize', 'de-slop', 'AI tells', or 'deep humanize'.
How do the two skills work together in practice?▼
Run the humanizer skill first to handle words and phrasing, then apply structural-humanizer as the second pass. This ordering is recommended because the StoryScope study (Russell et al. 2026) shows stylistic rewriting alone moved AI detection only 1.6 points, while narrative structure detects AI text at 93.2% F1.
What is the humanizer skill based on and who created it?▼
The humanizer skill (version 2.1.1) is based on Wikipedia's comprehensive 'Signs of AI writing' guide, covering inflated symbolism, superficial -ing analyses, negative parallelisms, and excessive conjunctive phrases. It credits the original skill by @blader at github.com/blader/humanizer.
What permissions or dependencies do these skills require?▼
The humanizer skill declares allowed-tools of Read, Write, Edit, Grep, Glob, and AskUserQuestion, meaning it operates directly on your text files within the skill runtime. No external services, accounts, or additional prerequisites are specified in the manifest.