AminBlg/SimpleEnglish

Agent skill: make LLMs write docs in ASD-STE100 Simplified Technical

What it solves

It eliminates "AI slop" and ambiguous phrasing in technical documentation by forcing LLMs to adhere to the ASD-STE100 Simplified Technical English (STE) standard. This prevents the common AI tendency to use corporate jargon, run-on sentences, and hedging language that can lead to dangerous misinterpretations in high-stakes environments like aerospace maintenance.

How it works

The project provides an agent skill (following the Agent Skills standard) and system prompts that implement 53 numbered rules from the STE standard. These rules enforce strict constraints on sentence length (max 20-25 words), verb tenses, voice (active only), and vocabulary consistency (one word = one meaning). It can be installed as a plugin for tools like Claude Code, Cursor, and VS Code Copilot, or pasted as a system prompt into any LLM interface.

Who it’s for

Technical writers, engineers, and developers who need to produce high-precision documentation, error messages, runbooks, incident reports, and release notes that are impossible to misread and easy to localize for non-native speakers.

Highlights

  • Broad Compatibility: Works with Claude Code, Cursor, Copilot, Gemini CLI, and other Agent Skills standard harnesses.
  • Measurable Improvement: Benchmarks show a 74.6% reduction in STE violations per 100 words across multiple models.
  • Detailed Rule Set: Implements specific constraints against "-ing" verb forms, hedging words (should/would/may), and complex sentence structures.
  • Versatile Application: Includes specific adaptations for error messages, runbooks, and system prompts to ensure they are treated as precise procedures.

Related

  • Dispatch
    SimpleEnglish Agent Skill Enables LLMs to Write in ASD‑STE100 Simplified Technical English

    The SimpleEnglish agent skill forces LLMs to produce documentation that complies with the ASD‑STE100 standard, cutting STE violations by 72.9% and shortening output across multiple Claude models.

  • Project
    danyuchn/asd-ste100-skill

    A Claude Code skill that rewrites ambiguous English into Simplified Technical English (STE) to ensure AI agents can parse instructions and tool descriptions without ambiguity.

  • Project
    coji/natural-japanese

    An Agent Skill for writing natural, professional Japanese by combining mechanical linting of AI-typical patterns with a structured writing constitution and iterative AI refinement.

  • Project
    HoangNguyen0403/agent-skills-standard

    A portable SDLC standards layer for AI coding agents that turns engineering rules into modular, version-controlled skills to ensure consistent code quality while reducing token waste.

  • Dispatch
    Improving LLM-Assisted Code Quality with agent.md

    A developer shares a system for using an agent.md file to enforce strict coding standards and commit message rules for AI agents, reducing the need for repetitive manual reviews.