agentleFS
nikolaypavlovnikolaypavlov/claude-skills/ste-writing/SKILL.md

ste-writing

Skill0 starsChanged 2 months agoMITView on GitHub

This skill should be used when the user wants prose rewritten into a controlled plain technical style to remove AI slop - documentation, READMEs, PR descriptions, error messages, release notes, comments (never code). English follows ASD-STE100 Simplified Technical English. Ukrainian follows adapted STE rules plus DSTU 3966:2009 style guidance. Triggers (English or Ukrainian): - "make writing not sound like AI", "make docs clear or plain" - "enforce controlled writing style", "plain technical English" - "write technical documentation that reads human" - "перепиши простіше", "спрости текст" - "прибери ШІ-стиль", "зроби текст людським" - "технічний стиль", "перепиши по-людськи" Two modes - strict (procedures, safety text) and STE-flavored (general prose).

13 other projects use a version of this, 3 of them edited.

Adopted by

Most recently taken up by

Used with

Projects with it also use prototype and grilling.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • GoogleSkill

    prototype

    Build a throwaway prototype to answer a design question. Use when the user wants to sanity-check whether a state model or logic feels right, or explore what a UI should look like.

    Also in 3 projects that use this

  • GoogleSkill

    grilling

    Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 13 edited versions we read keep these.

  • 4

    Distinctive

    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:

    Write prose in a controlled plain technical style.

    line 23
  2. Distinctive:

    Before you rewrite, read the ruleset for the target language:

    line 42
  3. Distinctive:

    Read the matching reference file.

    line 64
  4. Distinctive:

    Run the shared self-lint below, then the per-language checklist from the reference file.

    line 66

“Common” is measured: kept by at least 6 in 10 of the 13 edited versions of this file we could read. “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

ste-writing

Write prose in a controlled plain technical style. English text follows ASD-STE100 Simplified Technical English. Ukrainian text follows the same mechanical rules adapted to Ukrainian, plus DSTU 3966:2009 style guidance.

Scope

Apply this skill to documentation, READMEs, pull-request text, error messages, release notes, and comments. Do not apply it to code, identifiers, or command syntax. Do not use it for marketing copy, essays, or anything that needs a voice. The style strips voice on purpose.

Language selection

  1. Rewrite in the language of the source text.
  2. An explicit user instruction ("rewrite this in Ukrainian") overrides the source language.
  3. For mixed-language input, rewrite each part in its own language unless the user says otherwise.

Before you rewrite, read the ruleset for the target language:

  • English: references/english.md
  • Ukrainian: references/ukrainian.md

Modes

The same criteria apply to both languages:

  • strict - procedures, runbooks, safety text, error messages: apply every rule and both sentence-length caps.
  • STE-flavored - general prose (READMEs, PR descriptions, docs): apply the sentence, paragraph, and active-voice discipline; relax the dictionary lockdown so the text keeps enough range to read naturally.

Pick strict when the text tells a person what to do. Pick STE-flavored when the text explains or describes. The user can force either mode.

Workflow

  1. Detect the language of the source text.
  2. Pick the mode.
  3. Read the matching reference file.
  4. Rewrite the text.
  5. Run the shared self-lint below, then the per-language checklist from the reference file.
  6. Output only the requested text. No preamble, no summary, no closing remarks.

Self-lint (shared, language-neutral)

  1. Any sentence over the length cap? Split it.
  2. Any semicolon? Write two sentences.
  3. Any passive voice with a known actor? Make it active.
  4. Any noun chain where a verb works ("perform an analysis")? Use the verb.
  5. Same thing named two ways? Pick one name.
  6. Any banned filler or marketing phrase? Delete or replace it.

Each reference file adds language-specific checks: contractions and American spelling for English; calques, active participles, and -ся passives for Ukrainian.

Limits

The mechanical rules are lintable and remove the form of slop. Full STE also needs human judgment: the right technical noun, whether a sentence makes good sense. A checker cannot certify that. This skill fixes the FORM of slop. It cannot make a hollow paragraph true.

The official ASD-STE100 standard is free but copyrighted. Do not paste it in full: https://asd-ste100.org

Reference files

  • references/english.md - full ASD-STE100 ruleset: words, verbs, sentences, punctuation, structure, mode deltas, English self-lint.
  • references/ukrainian.md - adapted ruleset for Ukrainian plus DSTU 3966:2009 guidance: calques, participles, -ся passives, -но/-то forms, Ukrainian self-lint.

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.