writing-simply

Rewrites prose in ASD-STE100 Simplified Technical English and scores it with a linter.

1|Updated Feb 10, 2026
One-click install
npx skills add https://github.com/stefanfaur/roach-marketplace --skill writing-simply-stefanfaur
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-simply
Source: https://github.com/stefanfaur/roach-marketplace/tree/main/roach/skills/writing-simply
Command: npx skills add https://github.com/stefanfaur/roach-marketplace --skill writing-simply-stefanfaur

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? AI-generated and human-drafted prose often carries filler, passive voice, marketing adjectives, and inflated phrasing that readers must wade through. This Skill rewrites documentation, READMEs, changelogs, PR descriptions, and error messages in ASD-STE100 Simplified Technical English, a controlled language built for unambiguous technical prose, and verifies the result with a deterministic linter. ## Core Features & Use Cases - Rule-based rewriting: Applies STE rules covering word choice, active voice, sentence length caps (20 words for instructions, 25 for descriptions), punctuation, and paragraph structure. - Two modes: Strict mode for procedures, runbooks, and safety text; STE-flavored mode for general prose like READMEs and PR descriptions. - Deterministic linting: Ships ste-lint.py, which scores violations per 100 words across categories like passive voice, nominalizations, banned words, marketing adjectives, and long sentences, and counts em dashes as an AI-slop marker. - Use Case: You draft a README that sounds like generic AI output. Run the draft through the rewrite rules, then lint it with uv run ste-lint.py README.md, apply fixes, and lint again to confirm the score dropped. ## Quick Start Rewrite the attached README draft in plain Simplified Technical English and then lint it with ste-lint.py to report the violation score.

Frequently Asked Questions about writing-simply

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I remove AI slop from my writing?▼

Rewrite the text using ASD-STE100 Simplified Technical English rules: active voice, short common words, one instruction per sentence, and no marketing adjectives. Then run ste-lint.py on the draft to get a violations-per-100-words score and iterate until it drops.

What is ASD-STE100 Simplified Technical English?▼

ASD-STE100 is a controlled language written for aircraft maintenance manuals where a misread sentence can cause real harm. It enforces one name per thing, approved vocabulary, active voice, and strict sentence length limits, which strips out the voice patterns associated with AI-generated prose.

How do I check if my text sounds like AI wrote it?▼

Run ste-lint.py on the file or pipe the text through stdin. It reports violations per 100 words for passive voice, nominalizations, banned words, and marketing adjectives, plus a separate em dash count, which is a common AI prose marker.

When should I not use Simplified Technical English?▼

Avoid STE for prose that needs a voice: essays, marketing copy, conference talks, and persuasive writing. STE strips voice on purpose, so it removes the quality those formats depend on. Use it for documentation, runbooks, and error messages instead.

What is the difference between strict and STE-flavored mode?▼

Strict mode applies every rule and both sentence length caps, and suits procedures, runbooks, and safety text. STE-flavored mode keeps the sentence, paragraph, and active-voice discipline but relaxes the dictionary lockdown, suiting READMEs and PR descriptions.

Can the linter guarantee my text fully complies with STE?▼

No. The linter covers only the machine-checkable subset of STE, such as sentence length and banned words. Full compliance also requires judgment a checker cannot supply, like choosing the correct technical noun or verifying a sentence makes sense.