Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Human voice test - Read first 3 paragraphs aloud. Would someone guess AI wrote this? If yes, rewrite with more variety and imperfection
Component density - Count all components (QueryAnswer, TL;DR, ExpertInsight, EngagementHook, etc). Max 5-7 (standard), 8-12 (pillar)
Intentional typos - Count typing errors. Need 2-4 per article (thier, recieve, definately). NOT grammar errors
AI pattern detection - Check for: All answer-first paragraphs? Transition words (Moreover, Furthermore)? Meta-commentary (It's important to note)? If yes, fix
⭐ Readability Scores (Every Article)
Use hemingwayapp.com or readabilityformulas.com:
Hemingway Grade: 6-8 (max 10 acceptable)
Flesch Reading Ease: 60-70 (Plain English)
Flesch-Kincaid Grade: 8-10 (max 12 for technical)
Sentence length: <20% of sentences exceed 25 words
ExperienceBlock data - Real case studies OR marked as placeholder. NO fabricated data
Author credentials - Verifiable via LinkedIn profile link
All statistics sourced - No "Studies show..." without citation and link
🎯 Information Gain (High-Value Articles)
At least ONE of: (1) Original data/survey, (2) Unique case study, (3) Contrarian insight backed by evidence, (4) New framework/methodology, (5) Expert quote from interview
Output Required
Confirm:
All ⭐ checks passed - List any failures that need fixing