from humanizalo50
Detects 40 AI writing ‘tells’ and rewrites text to inject a distinct human voice, with a 3-pass self-audit loop and 6‑dimension scoring.
Humanizalo inspects prose for 40 specific AI-generated patterns across vocabulary, structure, formatting, and communication artifacts, removes those tells, and rewrites the text to sound like a specific human voice. It runs a multi-pass audit loop: draft rewrite, self-interrogation listing remaining tells, and a final rewrite until a 6-dimension scoring threshold is met (or three iterations complete). The skill outputs a humanized draft, a score breakdown, and a change summary.
Use Humanizalo when you need to convert machine-like output (reports, blog drafts, marketing copy, or translations) into convincing, personality-driven human prose. It’s appropriate for editing AI drafts before publishing, crafting author voices, or performing quality control on generated content.
Designed for agent environments that support rich instruction-following (Claude Code, Gemini/CLI-style agents, Copilot-like assistants); relies on tools: Read, Write, Edit, Grep, Glob, AskUserQuestion.
This skill has not been reviewed by our automated audit pipeline yet.
