Skip to content
← Tools / Skills
Skill · .md packageAminBlg/10 trustMetrics-checked
SimpleEnglish logo

SimpleEnglish

Force LLMs to write docs in ASD-STE100 Simplified Technical English — measured 73% fewer violations.

Visit official site →

About

SimpleEnglish is an agent skill that makes LLMs write documentation in ASD-STE100 Simplified Technical English, the controlled language aerospace has used since 1983. The author benchmarked it across 6 Claude models and 96 runs and reports 73% fewer STE violations per 100 words, with shorter output.

It ships adaptations for error messages, runbooks, incident reports, release notes, and AGENTS.md files. Limitation: the output reads like an Airbus manual — flat and impossible to misread, which is the point for docs but wrong for marketing or voice-driven writing.

What it can do

  • STE-constrained technical writing
  • Error message rewriting
  • Runbook authoring
  • Incident report drafting
  • Release notes
  • Prompt/system-prompt hardening

Setup

  1. Run: npx skills add AminBlg/SimpleEnglish
  2. Pick your agents in the installer prompt.
  3. Ask for any technical writing, or say 'rewrite this with simple-english'.

Prerequisites

Any agent harness that reads the Agent Skills open standard.

Use cases

others

Works with

Claude CodeCursorCodexCopilotgemini-cli

Trust score — 6/10

/10 trustMetrics-checked

Scored 6/10 (metrics tier, metrics-v1). Evidence: 0 stars on the fork (AminBlg upstream); measured benchmark: 96 runs, 73% fewer violations; 9 commits, TDD-style evals; niche but real evidence. License: MIT.

★ 0 GitHub stars

How the trust score is built

Full methodology →
Adoption · stars/downloads, log-scaled vs peers
3.0/10
Maintenance · recency, releases, issue load
5.0/10
Openness · license + docs
8.0/10
Manual review · sanity read + red-flag check
7.0/10
  • · 0 stars on the fork (AminBlg upstream)
  • · measured benchmark: 96 runs, 73% fewer violations
  • · 9 commits, TDD-style evals
  • · niche but real evidence

Sources

Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.