SimpleEnglish bets a spec, not a prompt, cures AI slop; 3.2k stars

SimpleEnglish is a free, MIT-licensed agent skill that makes LLMs write plain English with the discipline of ASD-STE100 Simplified Technical English, the controlled language aerospace has used since 1983: a Plain register by default (prose-only replies, five sentences max, no em-dashes, no openers or closers) plus a Strict mode that adds the STE dictionary, installable in Claude Code, Codex, Cursor, Copilot, Gemini CLI and about 25 more agents.

That an agent skill encoding ASD-STE100 will beat 'write clearly' prompts: agents follow numbered specs, and STE is a near-perfect negative of every AI writing tell.

AminBlg created the repository on 2026-07-21 around one line: 'your AI writes like a LinkedIn post — make it write like a Boeing manual.' The skill was built against the primary ASD-STE100 Issue 9 text (January 2025), not summaries, because agents without the skill invented rule numbers. The documented rules mirror the standard's numbered constraints: chat replies are prose-only with five sentences maximum, the first sentence answers, and em-dashes, openers and closers are banned; documents cap instructions at 20 words and descriptions at 25, put the condition before the command, and forbid should, would, may and might.

The project iterated publicly from 1.3.0 to 2.0.1 across 77 commits, with every benchmark reproducible from committed raw files. A self-audit dated 2026-09-02 (WHY-USELESS) found that the earlier headline — 81.3% fewer linter violations across 144 generations — measured rule obedience rather than what a reader sees, so 2.0.1 was rebuilt on reader-visible counts: over-cap sentences, em-dashes, bold, headers and bullets pooled over 16 claude-sonnet-4-6 replies fell from 406 to 58, and document linter violations per 100 words fell from 4.09 to 0.91 (78%). A community audit (issue #4) checked the vocabulary tables against the STE dictionary, and the fix was A/B-tested before it merged.

No company, funding or revenue is documented in the material; as of the 2026-09-04 crawl the project is live at 3.2k stars with v2.0.1 shipping, reproducible evals, and install paths for Claude Code, Codex and other agents.

Concrete traction: created 2026-07-21, the repo showed 3.2k stars (3,177 in crawl metadata), 112 forks and 77 commits at the 2026-09-04 crawl.

The bet is explicit and falsifiable: a numbered standard (ASD-STE100) should beat 'write clearly', and the README states why — 'Clearly' is an opinion, a word cap is a spec, and agents follow specs.

The project audits itself: an internal WHY-USELESS review found the old 81.3% linter metric measured rule obedience, not reader experience, and v2.0.1 was rebuilt on reader-visible defect counts.

Reproducible evidence: every benchmark ships committed raw files and scoring scripts, and community audit #4 checked the vocabulary tables against the STE dictionary before its fix merged.

If 'write clearly' fails, encode the standard: agents follow numbered specs, and the metric that counts is what a reader sees, not a rule-obedience score.

As of 2026-09-05 the repository is live at 3.2k stars (3,177 in crawl metadata), 112 forks and 77 commits, with v2.0.1 shipping. The README documents two registers — Plain (default) and Strict — installable via the skills CLI, as a Claude Code plugin with an output style, or as a Codex plugin with session hooks, or as a paste-in system prompt. Benchmarks dated 2026-09-02 and the WHY-USELESS self-audit are committed to the repo. No company, funding or revenue appears in the material, so SimpleEnglish remains a solo open-source bet whose commercial outcome is not documented.