Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when writing prose humans will read—documentation, commit messages, error messages, explanations, reports, or UI text. Applies Strunk's timeless rules for clearer, stronger, more professional writing.
.claude/skills/bilal140202-writing-clearly-and-concisely/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-04 | ✗→✓ | ▲ Improved | 59% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 9% | 0% |
| case-10 | ✓→✗ | ▼ Worse | 82% | 0% |
| case-05 | ✓→✓ | ＝ Same ✓ | 64% | 0% |
| case-01 | ✓→✓ | ＝ Same ✓ | 66% | 0% |
Write with clarity and force. This skill covers what to do (Strunk) and what not to do (AI patterns).
Use this skill whenever you write prose for humans:
If you're writing sentences for a human to read, use this skill.
When context is tight:
Loading a single section (~1,000-4,500 tokens) instead of everything saves significant context.
William Strunk Jr.'s The Elements of Style (1918) teaches you to write clearly and cut ruthlessly.
Elementary Rules of Usage (Grammar/Punctuation):
Elementary Principles of Composition:
The rules above are summarized from Strunk's original text. For complete explanations with examples:
| Section | File | ~Tokens | |---------|------|---------| | Grammar, punctuation, comma rules | 02-elementary-rules-of-usage.md | 2,500 | | Paragraph structure, active voice, concision | 03-elementary-principles-of-composition.md | 4,500 | | Headings, quotations, formatting | 04-a-few-matters-of-form.md | 1,000 | | Word choice, common errors | 05-words-and-expressions-commonly-misused.md | 4,000 |
Most tasks need only 03-elementary-principles-of-composition.md — it covers active voice, positive form, concrete language, and omitting needless words.
LLMs regress to statistical means, producing generic, puffy prose. Avoid:
Be specific, not grandiose. Say what it actually does.
For comprehensive research on why these patterns occur, see signs-of-ai-writing.md. Wikipedia editors developed this guide to detect AI-generated submissions — their patterns are well-documented and field-tested.
Writing for humans? Load the relevant section from elements-of-style/ and apply the rules. For most tasks, 03-elementary-principles-of-composition.md covers what matters most.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-05 | pass→pass | 5,821 | 4,235 | -27% | 1 | 1 | 0% | 985 | 1,620 | +64% | 0 | 0 | — |
case-01 | pass→pass | 5,752 | 4,896 | -15% | 1 | 1 | 0% | 1,000 | 1,660 | +66% | 0 | 0 | — |
case-02 | pass→pass | 6,439 | 5,374 | -17% | 1 | 1 | 0% | 1,236 | 1,867 | +51% | 0 | 0 | — |
case-03 | fail→fail | 6,528 | 4,455 | -32% | 1 | 1 | 0% | 1,160 | 1,605 | +38% | 0 | 0 | — |
case-04 | fail→pass | 7,689 | 4,791 | -38% | 1 | 1 | 0% | 1,094 | 1,739 | +59% | 0 | 0 | — |
case-06 | fail→fail | 4,799 | 4,274 | -11% | 1 | 1 | 0% | 857 | 1,701 | +98% | 0 | 0 | — |
case-07 | fail→fail | 7,245 | 5,691 | -21% | 1 | 1 | 0% | 1,055 | 1,734 | +64% | 0 | 0 | — |
case-08 | pass→pass | 6,284 | 5,429 | -14% | 1 | 1 | 0% | 1,062 | 1,810 | +70% | 0 | 0 | — |
case-09 | pass→pass | 5,706 | 2,822 | -51% | 1 | 1 | 0% | 913 | 1,322 | +45% | 0 | 0 | — |
case-10 | pass→fail | 6,040 | 4,342 | -28% | 1 | 1 | 0% | 890 | 1,622 | +82% | 0 | 0 | — |
case-11 | fail→pass | 8,214 | 3,420 | -58% | 1 | 1 | 0% | 1,304 | 1,425 | +9% | 0 | 0 | — |
case-12 | pass→pass | 6,131 | 4,090 | -33% | 1 | 1 | 0% | 1,045 | 1,541 | +47% | 0 | 0 | — |
case-13 | pass→pass | 4,306 | 2,775 | -36% | 1 | 1 | 0% | 689 | 1,318 | +91% | 0 | 0 | — |
case-14 | pass→pass | 3,654 | 2,540 | -30% | 1 | 1 | 0% | 540 | 1,266 | +134% | 0 | 0 | — |
case-15 | pass→pass | 4,498 | 5,960 | +33% | 1 | 1 | 0% | 849 | 1,831 | +116% | 0 | 0 | — |
case-16 | pass→pass | 5,772 | 4,052 | -30% | 1 | 1 | 0% | 948 | 1,469 | +55% | 0 | 0 | — |
case-17 | pass→pass | 6,228 | 4,076 | -35% | 1 | 1 | 0% | 1,067 | 1,520 | +42% | 0 | 0 | — |
case-18 | pass→pass | 5,801 | 5,426 | -6% | 1 | 1 | 0% | 994 | 1,894 | +91% | 0 | 0 | — |
case-19 | pass→pass | 5,936 | 4,623 | -22% | 1 | 1 | 0% | 1,013 | 1,669 | +65% | 0 | 0 | — |
case-20 | pass→pass | 9,014 | 4,730 | -48% | 1 | 1 | 0% | 1,565 | 1,531 | -2% | 0 | 0 | — |
case-21 | pass→pass | 3,953 | 3,614 | -9% | 1 | 1 | 0% | 673 | 1,549 | +130% | 0 | 0 | — |
case-22 | pass→pass | 5,159 | 2,999 | -42% | 1 | 1 | 0% | 884 | 1,396 | +58% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +5 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.