natural-copy-editing
Skill3 starsChanged 3 months agoMITView on GitHub
Use when the user asks to translate, correct English, polish a message, adapt tone, or restructure text and wants clean copy-paste output. For removing AI-sounding patterns, use humanizer.
In this repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agentfolio-operatorSkill
- branch-portSkill
- code-commentsSkill
- code-namingSkill
- cognitive-refactorSkill
- memory-palaceSkill
- testing-business-rulesSkill
- thorough-pr-reviewSkill
The file
We split this skill into the rules it gives an agent
12 rules. About 0 are common, 11 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈11
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
If the text is casual, keep it casual.
- Distinctive: line 16
If the text is professional, make it clear without sounding corporate or robotic.
- Distinctive: line 17
Prefer natural, direct writing.
- Very specific: line 18
If the text sounds machine-written, apply the humanizer pattern list.
- Distinctive: line 19
Avoid corporate filler like "I hope this message finds you well" and "kindly" unless it matches the user's tone.
- Distinctive: line 20
Use straight apostrophes only: I'm, don't, it's, you're.
- Distinctive: line 21
Do not use em dashes. Use a hyphen "-" when a dash is needed.
- Distinctive: line 26
Do not add labels like "Improved version:" unless requested.
- Distinctive: line 27
Do not provide multiple options unless the user asks for options.
- Distinctive: line 28
Do not wrap the answer in quotation marks or code fences unless requested.
- Distinctive: line 29
Do not format as Markdown unless the user explicitly asks for Markdown, or says it is for a Markdown-like platform such as a GitHub pull …
- Distinctive: line 30
Keep formatting minimal. Avoid excessive headings, bold text, bullets, tables, or decorative structure unless requested.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Natural Copy Editing
Core Rule
Assume the user wants clean copy-paste output. Give the revised text, not commentary about the revision, unless they ask for explanation or options.
Defaults
- Preserve the user's original tone unless they request a different one.
- If the text is casual, keep it casual.
- If the text is professional, make it clear without sounding corporate or robotic.
- Prefer natural, direct writing. Remove all mannered prose.
- If the text sounds machine-written, apply the
humanizerpattern list. Keep this skill's output format: final text only, no critique. - Avoid corporate filler like "I hope this message finds you well" and "kindly" unless it matches the user's tone.
- Use straight apostrophes only: I'm, don't, it's, you're.
- Do not use em dashes. Use a hyphen "-" when a dash is needed.
Output Format
- For direct rewrite, translation, and correction requests, output only the revised text.
- Do not add labels like "Improved version:" unless requested.
- Do not provide multiple options unless the user asks for options.
- Do not wrap the answer in quotation marks or code fences unless requested.
- Do not format as Markdown unless the user explicitly asks for Markdown, or says it is for a Markdown-like platform such as a GitHub pull request, Linear ticket, Notion, Obsidian, README, issue, or documentation.
- Keep formatting minimal. Avoid excessive headings, bold text, bullets, tables, or decorative structure unless requested.
Translation And English Corrections
- For translations, preserve the intended tone and context rather than translating word-for-word.
- For translations, output only the translated text unless the user asks for notes.
- For English corrections, fix grammar, clarity, flow, and structure while preserving the user's voice.
- Preserve meaning, urgency, hedging, politeness, and constraints like "no rush" or "today".
Common Mistakes
| Mistake | Fix |
| Adding "Here's a cleaner version" | Output only the revised text |
| Making casual text corporate | Keep it natural and close to the user's voice |
| Offering three versions unasked | Give one strong version |
| Explaining each change | Only explain if asked |
| Using Markdown by default | Use plain text unless requested |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

