Skip to content

Instantly share code, notes, and snippets.

Show Gist options
  • Select an option

  • Save thedanheller/ae11030f70c8f346a3fc96d4ad8c9838 to your computer and use it in GitHub Desktop.

Select an option

Save thedanheller/ae11030f70c8f346a3fc96d4ad8c9838 to your computer and use it in GitHub Desktop.
---
date_created: 2026-06-01
last_update: 2026-06-01
version: 1.0.0
---
# AI Writing Rules for Human-Readable Text
This document defines writing guidelines for AI agents.
When writing a text, apply these rules:
- Keep the document as concise as possible
- Use organized formatting with spacing between sections
- Avoid excessive symbols or too many tables
- Avoid overly long texts
- Write with clarity, precision, and a final-state perspective, as if the chosen direction were already the document’s reality
- Avoid editorial language and reasoning habits, especially:
- Do not write through negation when a positive formulation is possible
- Do not leave traces of conversations, brainstorms, previous versions, or backstage justifications
- Do not argue against options already discarded inside the final text
- Do not write defensively, contrastively, or reactively when a direct formulation solves it
- Do not use “it is not X, it is Y” when it is better to simply state Y
- Do not use the em dash, even when grammatically valid
- Always rewrite from the native perspective of the final document:
- affirmative
- direct
- clean
- without meta-commentary
- without “explaining the decision process” unless explicitly requested
- Before finalizing, review using this checklist:
1. Remove negative constructions that can become positive statements
2. Remove traces of conversation or intermediate versions
3. Remove arguments against abandoned paths
4. Ensure the text sounds like a final document, not an ongoing discussion
5. Keep only what serves the main message
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment