AIAny
AI Agent2026
Icon for item

SimpleEnglish

Applies concrete plain-English rules to AI-generated technical documents and chat replies. It defaults to layman-readable writing, with stricter ASD-STE100 vocabulary rules on request; it does not certify compliance.

Introduction

A vague request to “write clearly” gives a language model little to follow. SimpleEnglish takes a more testable approach: it turns clarity into rules for sentence length, voice, word choice, and structure, drawing on a controlled language used in aerospace maintenance writing. The useful distinction is that plain English is the default, while stricter vocabulary checks are reserved for users who explicitly request STE-style compliance.

What Sets It Apart
  • Separate rules guide chat replies and longer documents, so concise answers do not inherit the same constraints as procedures or descriptions.
  • Conditions come before commands, and each word should keep one meaning across a document. These rules target common sources of confusion, not just long sentences.
  • The project reports fewer formatting defects and fewer STE-linter violations in tests of version 2.0.1. Those results do not establish how later versions perform, and the linter measures rule compliance rather than reader comprehension.
  • Strict mode is not certification. ASD does not certify writing tools, and the project does not include the standard’s restricted vocabulary lists.
Who It’s For

Great fit if you use an AI agent to draft technical documentation for readers outside your field, especially procedures, runbooks, or error messages. Look elsewhere if you need marketing copy, a particular brand voice, or guaranteed ASD-STE100 compliance. The rules deliberately favor plain, restrained prose over persuasive style.

Information

  • Websitegithub.com
  • AuthorsAminBlg
  • Published date2026/07/21

Categories

More Items

GitHub
AI Agent2026

Turns research questions into reports through staged search, adversarial review, and citation checks, while saving sources in a persistent searchable vault. Runs with Claude Code or OpenAI Codex; full research can take hours.

GitHub
AI Agent2024

Runs research, coding, and document tasks inside its own Linux workspace, using a terminal, browser, and desktop apps. Docker isolation and the option to step in make its computer-level actions more manageable.

GitHub
AI Agent2023

Delegate coding, debugging, and refactoring tasks to an AI agent from the terminal using natural language. GitHub integration brings repository context into the workflow, with actions available for review before execution.