Back to list
pe200012

avoiding-ai-writing

by pe200012

0🍴 0📅 Jan 19, 2026

SKILL.md


name: avoiding-ai-writing version: 1.0.0 description: Use when writing wiki-style articles, documentation, summaries, or creative content to prevent "AI accent", puffery, and formulaic structures. author: pe200012 license: BSD-3-Clause tags: [writing, ai, communication, documentation]

Avoiding AI Writing

Overview

Writing like a human means embracing irregularity, neutrality, and concrete detail over formulaic structure and dramatic puffery.

AI models naturally gravitate towards "high school essay" structures: generic intros, numbered lists for body paragraphs, and moralizing conclusions. They overuse specific "smart-sounding" words (delve, testament, tapestry). This skill forces a "Wikipedia-neutral" or "Journalistic" tone by banning these patterns.

When to Use

  • Writing Wikipedia-style articles or encyclopedic entries
  • Creating documentation or summaries that need to sound professional, not "generated"
  • When user asks for "natural", "human-like", or "non-AI" writing
  • Symptoms of failure: Output contains "In conclusion", "Ultimately", numbered lists for "Impact", or words like "burgeoning", "testament", "tapestry".

The "AI Accent" Red Flags (STOP & FIX)

If you see these, delete and rewrite:

CategoryThe "AI Accent" (BANNED)Human/Neutral Replacement
Puffery"stands as a testament to", "serves as a reminder", "beacon of hope", "rich tapestry""shows", "demonstrates", "is", [delete entirely]
Connectors"It is important to note", "It is worth mentioning", "underscoring the significance of"[Just state the fact], "This implies..."
Vocabulary"delve", "realm", "landscape", "foster", "spearhead", "burgeoning", "crucial role""investigate", "field", "area", "encourage", "led", "growing", "role"
StructureNumbered lists for "Legacy" or "Impact"; Title Case headersProse paragraphs; Sentence case headers
Closers"Ultimately,", "In conclusion,", "looking forward,"[End on a specific fact or event], [No conclusion needed]

Core Rules

1. Flatten Lists into Prose

AI Instinct: Create a numbered list for "Key Features" or "Legacy". Human approach: Weave these points into cohesive paragraphs.

2. Kill the "Conclusion Paragraph"

Encyclopedic entries do not need a "wrap up" moralizing the subject.

  • Delete any paragraph starting with "Ultimately," "In summary," or "Today, X stands as...".
  • End with the last chronological event or a specific status (e.g., "The site was designated a landmark in 1999.").

3. Neutralize "Significance"

Don't tell the reader something is significant using abstract praise. Show the significance using facts.

  • ❌ "Her work served as a testament to the power of activism."
  • ✅ "Her lobbying led to the 2008 constitutional amendment acknowledging nature's rights."

4. Sentence Case for Headers

Wikipedia uses Sentence case for headers (only first word capitalized), NEVER Title Case.

  • ## Early Life and Education
  • ## Early life and education
  • ## Impact on Watchmaking
  • ## Impact on watchmaking

Refactoring Process (The "De-bot" Cycle)

  1. Draft the content.
  2. Scan for "delve", "testament", "landscape", "underscoring". Destroy them.
  3. Check Structure: Is there a numbered list? Flatten it.
  4. Check End: Is there a moral summary? Delete it.
  5. Check Headers: Are they Title Case? Lowercase them (except proper nouns).

Example: Refactoring a Biography

Legacy

Her legacy is defined by three pillars:

  1. Legal Pioneer: She spearheaded the "Rights of Nature" movement.
  2. Corporate Accountability: She fostered a new era of responsibility.

Ultimately, Mora serves as a beacon of hope for future generations.

Legacy

Mora is credited with establishing the legal framework for the "Rights of Nature" argument used in Ecuador's 2008 constitution. Her tactics shifted environmental activism from street protests to shareholder lobbying, establishing early precedents for corporate environmental accountability. Modern climate organizations, including the Youth Climate Coalition, cite her strategies as foundational to their 2020 campaigns.

Score

Total Score

50/100

Based on repository quality metrics

SKILL.md

SKILL.mdファイルが含まれている

+20
LICENSE

ライセンスが設定されている

+10
説明文

100文字以上の説明がある

0/10
人気

GitHub Stars 100以上

0/15
最近の活動

3ヶ月以内に更新がある

0/10
フォーク

10回以上フォークされている

0/5
Issue管理

オープンIssueが50未満

+5
言語

プログラミング言語が設定されている

+5
タグ

1つ以上のタグが設定されている

0/5

Reviews

💬

Reviews coming soon