// AI TEXT CLARITY & CONTROLLED-LANGUAGE REWRITING ALTERNATIVE

STE80 Humanizer Alternative: Govern Clarity at Generation and Publish It, Instead of Linting Text After

An STE80 Humanizer alternative: compare STE80 vs Kompozy — where the free ASD-STE100 clarity skill fits, where it stops, and when to generate instead.

Last verified · 2026-11-10 · by Moe Ameen

STE80 Humanizer and Kompozy both come up when you want AI writing that reads clean, but they sit at opposite ends of the pipeline and barely overlap. STE80 is a free, open-source agent skill that rewrites a block of existing English toward ASD-STE100 clarity — short sentences, one idea at a time, active voice — inside your coding agent. Kompozy is a content generation and publishing engine: it writes copy in your voice in the first place, then turns that copy into video, images, carousels, blogs, and newsletters and publishes them across platforms.

I want to be fair up front, because STE80 is genuinely good and genuinely honest. It refuses the usual "humanizer" job — it explicitly does not try to beat AI detectors, it optimizes for a reader's comprehension — and it is free under the MIT license. If your entire need is "make this paragraph clearer," STE80 is the right, cheaper tool, and you do not need this page. The reason someone searches for an alternative is usually one of two things: they want clarity without living in a coding agent, or they realize a text-only linter does nothing about the actual job of producing and shipping on-brand content.

That is the comparison this page makes. STE80 polishes one block of text after it is written. Kompozy enforces a clean, controlled style at generation time — a Persona Brief fixes the voice and a banned-word filter strips the same AI tells STE80's first pass targets — and then keeps going into the formats and the multi-platform publishing a rewriter never touches. Prices and features below are reconciled against STE80's public repository and Kompozy's pricing, checked 2026-11-10.

What STE80 Humanizer does

STE80 Humanizer is an open-source "agent skill" published on GitHub by the account Dodaios under the MIT license. Rather than a website, it installs into coding agents — Claude Code, Codex, Cursor, and Claude.ai/Desktop — and ships a dependency-free Python linter. It implements Andrej Karpathy's October 2, 2026 suggestion to prompt models to write "about 80% of the way to ASD-STE100 Simplified Technical English," the controlled English built for aircraft maintenance manuals. You give it AI-generated or rough text and it rewrites it to read more naturally without changing the meaning, in two passes: Pass 1 removes 26 "AI tells" drawn from Wikipedia's "Signs of AI writing" and Siqi Chen's MIT-licensed blader/humanizer catalog, and Pass 2 applies 15 ASD-STE100-inspired structure rules. Three modes (Humanize, Light, Strict STE) set the intensity, and it can match a voice from a sample. It paraphrases the ASD-STE100 rules and ships no dictionary, is not affiliated with ASD, states that "STE style" is not formal "STE compliance," and is explicit that detector evasion is not a goal. It edits one block of English text — it generates no images, carousels, or video, holds no brand voice across a batch, and has no scheduling or publishing.

Why people look for a STE80 Humanizer alternative

STE80 does one narrow thing well and is silent on everything a content creator needs after the words exist. There is no persona or brand-voice system that generates copy in your tone — it rewrites whatever you paste. There is no image, carousel, or video generation. There is no scheduler and no multi-platform publishing. It is English-centric (the STE structure rules are English-only), it requires a coding agent rather than offering a GUI, and by the author's own note it suits explanatory prose — fiction, speeches, and brand copy get only the AI-tell pass. None of that is a flaw; it is a scope, and STE80 is refreshingly honest about it. But a clarity linter is a step you apply after a draft, and it touches only the text. For anyone whose endgame is published, on-brand content across platforms rather than a single cleaner paragraph, enforcing clarity at generation time and then producing the media is the fix a rewriter can never be.

STE80 Humanizer vs Kompozy — feature comparison

FeatureSTE80 HumanizerKompozyNote
Rewrite existing text toward clarity (ASD-STE100 style)Yes — core productNoSTE80's whole reason to exist; not something Kompozy markets as a standalone rewriter.
Generates copy in your brand voiceNoYesKompozy's Persona Brief governs tone and vocabulary at generation time; STE80 rewrites text you supply.
Strips AI tells at generationPartialYesSTE80 removes tells by rewriting after the fact; Kompozy bans the tell-words during generation.
Controlled-language / readability structure rulesYes — 15 STE rulesPartialSTE80 applies explicit ASD-STE100-style structure; Kompozy follows a Persona Brief you can instruct toward the same clarity.
Video generation (avatar / clips / shorts)NoYesPersona Shorts, Persona HeyGen, Clipped Shorts, Marketing Shorts, Listicle Video and more.
Image generation (photos, carousels, quote cards)NoYesgpt-image scenes, face-locked persona images, multi-slide carousels, SVG quote graphics.
Blog + newsletter generationNoYesSTE80 only rewrites existing text; Kompozy writes brand blogs and email newsletters governed by the Persona Brief.
Multi-platform scheduling & publishingNoYesKompozy fans output to eight social platforms plus Mailchimp and blog from one queue.
InterfaceAgent skill (no GUI)Full web appSTE80 runs inside a coding agent; Kompozy is a hosted app with a review pipeline and calendar.
Language coverageEnglish-centricMulti-languageSTE structure rules are English-only; Kompozy generates in multiple languages.
Detector evasionNo — explicit non-goalNoNeither chases detector scores; STE80 optimizes readability, Kompozy optimizes on-brand generation.
PriceFree (MIT, open-source)PaidSTE80 is free; Kompozy is a paid generation-and-publishing engine. Different categories entirely.

Pricing — STE80 Humanizer vs Kompozy

TierSTE80 Humanizer planSTE80 Humanizer priceKompozy planKompozy price
EntrySTE80 Humanizer (open-source)Free — MIT license, no account, no dictionary to buyKompozy Starter$199/mo (5,500 credits across all buckets)
MidSTE80 Humanizer (open-source)Free — no paid tiers existKompozy Pro$499/mo (18,000 credits) — API, webhooks, workspaces
TopSTE80 Humanizer (open-source)Free — self-hosted in your agentKompozy EnterpriseContact Sales
Pricing verified 2026-11-10from each vendor’s public pricing page. Promotional rates rotate monthly — verify before purchase.

What STE80 Humanizer does well

  • Free and open-source under MIT — no signup, no subscription, no dictionary license.
  • Honest framing: optimizes for readability and explicitly disclaims beating AI detectors.
  • Genuinely improves expository prose with ASD-STE100-style structure rules.
  • Runs inside the coding agent you already use (Claude Code, Codex, Cursor, Claude.ai/Desktop).
  • Voice-matching from a sample and a fact-fidelity pass that flags source contradictions.
  • Transparent about limits — "STE style" is not formal "STE compliance," and the linter is a heuristic.
  • The underlying "80% of ASD-STE100" method is portable as a plain prompt.

Where STE80 Humanizer falls short

  • Does one narrow thing — rewrites a block of text — and generates nothing new.
  • STE structure rules are English-only; other languages get a reduced pass.
  • Requires a coding agent; no standalone web app or GUI for non-technical users.
  • Best on explanatory text; fiction, speeches, and brand copy get only the AI-tell pass.
  • Not a compliance tool — it will not produce certified ASD-STE100 for regulated docs.
  • No brand-voice governance across a batch, and no scheduling or multi-platform publishing.
  • Early-stage with lightly documented, self-described "hint, not proof" evals.

Pick STE80 Humanizer when…

  • You just need to make a block of text read clearer. If the job is "tighten this paragraph," a free controlled-language rewriter is the right shape of tool.
  • You already live in a coding agent. STE80 installs into Claude Code, Codex, Cursor, or Claude.ai/Desktop, so clarity edits happen where you already write.
  • You want zero cost and full control. MIT-licensed and open-source, with no dictionary to license and nothing to sign up for.
  • You are a developer or technical writer, not a publisher. For READMEs, docs, and explainers, a clarity pass is enough; Kompozy is a creator content engine you would not need.

Pick Kompozy when…

  • You publish content to social platforms. Kompozy generates posts, video, and images in your voice and publishes across eight social platforms plus blog and email — STE80 stops at rewritten text.
  • You want clean copy without running a second pass. A Persona Brief plus banned-word filter produces clear, on-brand copy at generation time, so there is nothing to lint afterward.
  • You need more than text. Carousels, quote cards, persona/avatar video, blogs, and newsletters all come from the same brief — a text rewriter makes none of these.
  • You want a calendar and a review gate, not a CLI. Kompozy fans one idea into many formats, routes each through a per-post review pipeline, and schedules them; STE80 polishes one block at a time in your editor.

Why Kompozy is the STE80 Humanizer alternative we recommend

Here is the honest split. STE80 and Kompozy are not substitutes — one is a free clarity linter for a paragraph, the other is a paid engine for producing and shipping content — and the best setup often uses both. Where they connect is the principle: STE80's best idea is that a clean, controlled style can be specified and applied, rather than hoped for. Kompozy does that at generation time instead of as a cleanup pass. The Persona Brief fixes your voice, you can instruct it toward the same clarity Karpathy described (one idea per sentence, short sentences, active voice), and a banned-word filter strips the exact AI tells STE80's first pass removes — so every draft comes out clean, not just the one block you remembered to run through a linter.

Then Kompozy does the part a text skill structurally cannot: one idea becomes captions, a blog, a newsletter, carousels, quote cards, and persona or avatar video, all in one voice, scheduled and published across the eight social platforms plus email and blog from a single queue behind a per-post review gate. So the choice is not "which clarity tool." It is "do I want to keep polishing text after the fact, or generate on-brand content that already reads clean and ships everywhere." If you only ever need a cleaner paragraph, STE80 is free and good — use it. If your real job is creating and publishing content, start where the content is made: Kompozy Starter is $199/mo (5,500 credits).

Frequently asked questions

Is Kompozy an alternative to STE80 Humanizer?

Only loosely — they are different categories. STE80 is a free skill that rewrites existing text toward ASD-STE100 clarity; Kompozy generates content in your brand voice from the start and publishes it across platforms. If your specific need is polishing a paragraph you already have, STE80 is the right tool and Kompozy is overkill. If you need to produce and ship on-brand content, Kompozy is the category you want.

STE80 Humanizer is free — why pay for Kompozy?

Because they buy different things. STE80 at $0 buys a clarity rewrite for existing English. Kompozy is paid because it generates finished media — video, images, carousels, blogs, newsletters — in a governed voice and publishes it across eight social platforms plus blog and email on a schedule. If all you need is cleaner text, stay free; if you need to create and distribute content, that is what you are paying for.

Does Kompozy follow ASD-STE100 or controlled-language rules?

Not as a formal standard, no — Kompozy is not an ASD-STE100 compliance tool, and neither is STE80 (it produces "STE style," not certified compliance). What Kompozy offers is a Persona Brief you can instruct toward the same readability habits — short sentences, active voice, one idea at a time — plus a banned-word filter that removes AI tells at generation time.

Can STE80 Humanizer create social posts or video?

No. STE80 rewrites a block of text you already have. It generates no images, carousels, or video, enforces no brand voice across formats, and has no scheduler or publishing. To turn clarified copy into multi-format content and publish it across platforms, pair it with — or replace it by — a content engine like Kompozy.

Is a controlled-language rewriter the best way to get readable content at scale?

For a one-off polish it is fine, and STE80 is a good free option. For ongoing content it is the harder path — you keep producing drafts and running them through a linter one block at a time. Generating clean, on-brand copy up front with a Persona Brief and banned-word filter, as Kompozy does, is more durable and scales into the media and multi-platform publishing a rewriter never touches.

Related deep guides

See Kompozy pricing · Get Started →