agentleFS
YassimbaYassimba/loom/skills/write-simply/SKILL.md

write-simply

Skill9 starsChanged 2 months agoApache-2.0View on GitHub

Writing or editing prose humans read - docs, commit messages, error messages, reports, UI text. Also the register other skills invoke for compact technical prose (docstrings, comments, captions, proposals), and the pass that makes text sound human instead of AI-written ("humanize", "reads like ChatGPT").

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Always use ASD-STE100 Simplified Technical English plus Zinsser's four principles Write vigorous prose: every word tells, sentences are …

    line 8
  2. Distinctive:

    Write vigorous prose: every word tells, sentences are active and concrete, needless words are gone.

    line 9
  3. Distinctive:

    Use the active voice.

    line 16
  4. Distinctive:

    Put statements in positive form.

    line 17
  5. Distinctive:

    Use definite, specific, concrete language.

    line 18
  6. Distinctive:

    Place emphatic words at the end of the sentence.

    line 21

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Write Simply

Always use ASD-STE100 Simplified Technical English plus Zinsser's four principles Write vigorous prose: every word tells, sentences are active and concrete, needless words are gone. Say what the thing does, in the words a knowledgeable human would pick.

Principles

Strunk's principles of composition, applied to everything you write:

  • One paragraph per topic; open it with a topic sentence.
  • Use the active voice.
  • Put statements in positive form.
  • Use definite, specific, concrete language.
  • Omit needless words.
  • Express co-ordinate ideas in similar form; keep related words together.
  • Place emphatic words at the end of the sentence.

Register for compact prose

Docstrings, comments, commit messages, captions, a numbered list of proposals: ASD-STE100 Simplified Technical English plus Zinsser's four principles — clarity, simplicity, brevity, humanity. On top of the principles above:

  • One idea per sentence; about 20 words or fewer.
  • One meaning per word; the same word for the same thing every time.
  • Simple present tense; instructions as imperatives.
  • Plain words; name the specific thing.
  • One person talking to another.

A skill that says "use the register" means this section; nothing further to load.

What to load

Match the task to one row and load exactly that file:

TaskLoad
Writing or editing paragraphs, docs, explanationsreferences/elements-of-style/03-elementary-principles-of-composition.md
Fixing grammar, commas, punctuation in existing textreferences/elements-of-style/02-elementary-rules-of-usage.md
Choosing the right word, fixing common misusesreferences/elements-of-style/05-words-and-expressions-commonly-misused.md
Headings, quotations, formattingreferences/elements-of-style/04-a-few-matters-of-form.md
load this alwaysreferences/humanizer.md
Spotting AI patterns while writing (words to watch)references/ai-patterns.md
Auditing whether text is AI-written (full catalogue)references/signs-of-ai-writing.md

Full pass — load 02, 03, 04, 05, humanizer.md, and ai-patterns.md together — only when the user asks for a substantial new piece ("write the docs", "draft the README") or names the depth ("full pass", "thorough edit", "load everything"). "Fix this paragraph" stays with one file.

Done means

Re-read the full output. It is done when every sentence passes:

  • each word informs;
  • the active voice wherever it works;
  • a specific where a generic stood;
  • the human's word where an AI pattern stood (puffery, promotional adjectives, "delve"/"leverage", decorative formatting).

A failed sentence is fixed, then the whole output is re-read again.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.