ste-writing
Skill3 starsChanged 39 days agoMITView on GitHub
Write or rewrite text in Simplified Technical English (STE) style. Use ONLY when the user explicitly asks for STE, Simplified Technical English, ASD-STE100, or "80% STE" writing, such as "explain X in STE" or "rewrite this in STE". Never use for ordinary answers.
13 other projects use a version of this.
Adopted by
Most recently taken up by
- tonkeeper/tonkeeper-webtheir own version, first seen Nov 9, 2026
- LeonardiFranco/skillstheir own version, first seen Nov 8, 2026
- xiduzo/wtftheir own version, first seen Nov 8, 2026
- vstorm-co/agenticostheir own version, first seen Oct 16, 2026
- berkeley-humanoids/skillstheir own version, first seen Oct 9, 2026
- berkeley-humanoids/Skillstheir own version, first seen Oct 9, 2026
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.
GoogleSkillprototype
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
GoogleSkillgrilling
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
Other versions13 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 13
- tonkeeper/tonkeeper-web★ 448same name
- vstorm-co/agenticos★ 241same name
- ahonn/dotfiles★ 62same name
- xiduzo/wtf★ 7same name
- berkeley-humanoids/skills★ 4same name
- berkeley-humanoids/Skills★ 4same name
- 1fc0nfig/ste-writing★ 2same name
- joshua-noel/skills★ 2same name
- TrimVis/skills★ 2same name
- HarveyZgit/agents★ 1same name
3 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .claude/CLAUDE.mdCLAUDE.md
- diagram-explainerSkill
- html-explainerSkill
- video-explainerSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
22 rules. 0 are common, 18 are distinctive, and 4 are very specific to this project.
0
Common
Most of the 13 edited versions we read keep these.
18
Distinctive
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Write in a controlled, easy-to-read style so the reader spends effort on the ideas, not on the sentences.
- Distinctive: line 20
Use active voice in instructions and in descriptions.
- Distinctive: line 22
Use one term for one concept.
- Distinctive: line 24
Write procedures as numbered lists, one step per item.
- Distinctive: line 26
Keep verb forms simple. Write "The server runs", not "The server is running".
- Distinctive: line 27
Keep articles ("the", "a", "an").
- Distinctive: line 31
Before: "Open the file, change the port, and save it."
- Distinctive: line 32
After: "1. Open the file. 2.
- Distinctive: line 40
Use strict only when the user asks for "strict STE" or "full STE".
- Distinctive: line 42
Use no progressive form ("is running") and no perfect form ("has finished").
- Distinctive: line 43
Use no passive voice in procedures.
- Distinctive: line 44
Keep noun clusters to 3 words or fewer.
- Very specific: line 45
Use no word or phrase from references/replacements.md.
- Distinctive: line 51
Leave these unchanged, exactly as written:
- Distinctive: line 62
Keep the style for every later prose answer until the user says to stop.
- Distinctive: line 66
When the answer is not in English, apply STE principles only:
- Very specific: line 74
Do not translate English rules word for word.
- Very specific: line 80
If Python is available and the draft is English, run the checker from this skill's scripts/ folder.
- Very specific: line 81
Read the JSON. Fix each sentence with over_limit true or a passive or progressive count above 0.
- Distinctive: line 82
Run the checker one more time.
- Distinctive: line 83
If Python is not available, the checker fails to run (permission denied, error), or the draft is not English, use this checklist on the …
- Distinctive: line 94
Never mention the checker or the self-check in the answer.
“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.
Files
STE writing
Write in a controlled, easy-to-read style so the reader spends effort on the ideas, not on the sentences. This skill is inspired by ASD-STE100. It is not official and is not endorsed by ASD.
There are two strictness levels. Relaxed is the default: the core rules below, in natural wording. A request for "80% STE" or "soft STE" means relaxed. English relaxed output carries no label or note about the style. Strict is only for a user who asks for "strict STE" or "full STE". It adds the rules in "Strict strictness" below. Full rules with before/after examples are in references/rules.md. Read it when a case is unclear.
Two uses
- Explain (main use): the user names a topic. Explain it with the rules below.
- Rewrite: the user gives text. Rewrite it with the rules below. Keep every fact. Do not add or remove meaning.
Core rules
- Procedural sentences: 20 words or fewer. Descriptive sentences: 25 words or fewer.
- Use active voice in instructions and in descriptions. Write "Restart the server", not "The server should be restarted".
- Give one instruction per sentence.
- Use one term for one concept. Do not switch between "folder" and "directory".
- Give each paragraph one topic. Use 6 sentences or fewer.
- Write procedures as numbered lists, one step per item.
- Mark hazards with WARNING (risk of injury) or CAUTION (risk of damage). Put the command first, then the risk.
- Keep verb forms simple. Write "The server runs", not "The server is running". The checker counts both forms.
- Keep articles ("the", "a", "an"). Do not drop them to save words. Write "Open the file", not "Open file".
Example of rule 3:
- Before: "Open the file, change the port, and save it."
- After: "1. Open the file. 2. Change the port. 3. Save the file."
Example of rule 7:
- CAUTION: Do not delete the
data/folder. The app loses all saved records.
Strict strictness
Use strict only when the user asks for "strict STE" or "full STE". Apply every core rule above, and also:
- Use no progressive form ("is running") and no perfect form ("has finished"). Use the simple form.
- Use no passive voice in procedures. Descriptions should stay active too.
- Keep noun clusters to 3 words or fewer. Write "the timeout of the connection pool", not "the connection pool timeout setting value".
- Use no word or phrase from
references/replacements.md. Use the plain choice in its second column. The list covers inflections ("utilize" also bans "utilizes" and "utilized").
End strict English output with ONE short note line: "Written in strict STE (not official ASD-STE100)." Put nothing else about the style in the answer. Relaxed output has no note.
Exempt content
Leave these unchanged, exactly as written:
- code and code blocks
- identifiers, commands, and error messages
- quoted source text
- proper names and product names
Apply the rules only to the prose around them.
Persistence
Keep the style for every later prose answer until the user says to stop. Answer a code-only follow-up normally. The style stays on for the next prose answer after it.
Non-English output
When the answer is not in English, apply STE principles only:
- short sentences
- active voice
- one instruction per sentence
- one term per concept
- one topic per paragraph
Do not translate English rules word for word. The strict-only rules and the replacement list do not apply. End the answer with one short note line, in the answer language, that says only STE principles were applied. See references/rules.md for an example.
Before you answer (self-check)
Measure the draft, then fix it. The checker in scripts/ste_check.py is the source of truth for the length limits and the counts.
- If Python is available and the draft is English, run the checker from this skill's
scripts/folder. Send the draft on stdin:python <skill folder>/scripts/ste_check.py --strictness relaxed --lang auto. Usepython3ifpythonis missing. For strict, use--strictness strict --replacements <skill folder>/references/replacements.md. - Read the JSON. Fix each sentence with
over_limittrue or apassiveorprogressivecount above 0. For strict, also fix eachperfectcount, each entry innoun_clusters, each entry inreplacement_hits, and each passive in a procedure. - Run the checker one more time. Fix what it still flags, then answer. Do not loop further.
- If Python is not available, the checker fails to run (permission denied, error), or the draft is not English, use this checklist on the draft. Fall back silently. Do not tell the user the checker failed:
- Is every sentence within its limit (20 procedural, 25 descriptive)?
- Does every instruction use active voice?
- Does every sentence hold one instruction?
- Does each concept use one term?
- English only: are the articles ("the", "a", "an") kept?
- Does each paragraph hold one topic, in 6 sentences or fewer?
- Are procedures numbered, and are hazards marked WARNING or CAUTION?
- Is exempt content unchanged?
- Strict only: no progressive or perfect forms, no passive in procedures, no noun cluster over 3 words, no word from
references/replacements.md, and one note line at the end?
Never mention the checker or the self-check in the answer. Show no check results, and report no failure to run the checker. The checker ignores exempt content.
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.

