> For the complete documentation index, see [llms.txt](https://docs.voicemoat.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.voicemoat.com/voice-and-audenai/style-and-stance.md).

# Style and Stance

**In short:** Studio checks drafts two ways. **Style** is how the draft looks next to your own posts: a count of your habits it matches. **Stance** is what the draft says: each line read against the claims, stories, and framings in your own posts. They are two separate results, never blended into one score. Style is free; Stance costs 1 credit when you run it.

#### Why two checks

A draft can match your sentence length, your emoji habits, and your openers, and still say something you would never say. Counting habits cannot see that. So VoiceMoat keeps the two apart: Style for how a draft looks, Stance for what it says.

#### How it fits together

| Layer           | What it answers                                                    | Where you see it                                       |
| --------------- | ------------------------------------------------------------------ | ------------------------------------------------------ |
| **Profile DNA** | Your voice model, built from your own content across ten signals   | Voice Lab                                              |
| **Style**       | Does the draft look like your content? Habits, AI tells, and drift | Every draft in Studio, free and live as you type       |
| **Stance**      | Does it say what you actually think?                               | Studio when you run it (1 credit), and every Grow post |
| **Feeling**     | What should the reader feel?                                       | The chip on every post audenAI writes                  |
| **Post Coach**  | Will it travel? Craft and reach tips that never change your voice  | Studio, in the Reach group                             |

There is no single voice score. VoiceMoat used to show one 0 to 100 Voice Match number. Style and Stance replaced it in October 2026, because one number could look high while a line said something you would never say.

#### Style

The Style card sits at the top of the **Voice** group in Studio's rail. It counts how many of your habits the draft matches, for example **4 of 6 checks match**, and sums it up as "Reads like your posts", "Close to how you write", or "Off from how you write".

| Check           | What it compares                                                                                                      |
| --------------- | --------------------------------------------------------------------------------------------------------------------- |
| Sentence length | How long your sentences usually run                                                                                   |
| Emoji use       | How often you use emoji                                                                                               |
| Casing          | Your capitalisation habits                                                                                            |
| Post length     | How long your posts usually are                                                                                       |
| Opener style    | The ways you actually open your posts                                                                                 |
| Filler words    | Filler words in the draft, such as "basically" or "honestly", that you don't use in your own posts                    |
| Line shape      | How many lines hold one sentence, against how often you put one sentence on a line                                    |
| AI tells        | Patterns that make a post read as AI, such as "it's not X, it's Y" or a bait question at the end                      |
| Drift           | Signs the draft is sliding away from you, such as fewer contractions, less "I", or longer words than you normally use |

Most checks that are off show **Yours**: a real line from your own posts, so you can see the difference instead of trusting a number. On Twitter, a reply is compared with your replies, not your posts.

**Filler words** and **Line shape** are new in October 2026. They come from the counts audenAI takes of your own posts (see The 10 signals), and they say what is different instead of quoting a line, for example "More one-sentence lines than you write." Both check posts only, because a reply is too short to judge, and Line shape waits until a draft has at least 4 lines.

Style is free and updates live as you type.

<figure><img src="https://708994200-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FpIMRJgVWZK5TkQX6SclK%2Fuploads%2Fgit-blob-92133488149aa017f6f2e3fa8ee4ddcacd6622b7%2Fstyle-card.png?alt=media" alt="The Style card: 5 of 6 checks match, reads like your posts. Drift is off, with a line from the creator&#x27;s own post next to the draft"><figcaption><p>The Style card. Drift is off here, so it shows a real line from the creator's own post next to what the draft does.</p></figcaption></figure>

#### Stance

Tap **Check stance** on the Stance card (1 credit). audenAI reads each line of the draft against your claim bank and gives one of three results:

| Result             | What it means                                                                         |
| ------------------ | ------------------------------------------------------------------------------------- |
| **On voice**       | Every line fits what you say.                                                         |
| **Check one line** | A line makes a new claim, or tells a story you haven't told. Not wrong, worth a look. |
| **Off voice**      | A line goes against a claim of yours, or uses a framing you avoid.                    |

Flagged lines are underlined in your draft. Each one shows what was found ("Goes against your claim", "A framing you avoid", "A story you haven't told", or "New claim"), a short reason, and the claim it hit, quoted from your post with a link to it.

<figure><img src="https://708994200-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FpIMRJgVWZK5TkQX6SclK%2Fuploads%2Fgit-blob-dcff02004cebdb97a26eb6a7e0386f270ac4f600%2Fstance-underline.png?alt=media" alt="A draft line underlined in red: So grind harder and post every day, no days off"><figcaption><p>The flagged line is underlined in the draft.</p></figcaption></figure>

<figure><img src="https://708994200-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FpIMRJgVWZK5TkQX6SclK%2Fuploads%2Fgit-blob-68b63b94cecd600378bf6da7e3ab9983633eaac6%2Fstance-card.png?alt=media" alt="The Stance card showing Off voice: the line uses a framing the creator avoids, quoted next to the claim it hit, with Suggest a rewrite, Ignore here and I say this now"><figcaption><p>Off voice: the line uses hustle talk, a framing this creator avoids. The claim it hit is quoted under it.</p></figcaption></figure>

What you can do with a flagged line:

* **Suggest a rewrite**: a rewrite from your own words, shown under the line. **Use this** puts it in your draft; **Keep mine** leaves your line as it is.
* **Ignore here**: hides it for this draft.
* **This is me now**: adds the line to your claim bank. If it updates a claim you already have, it asks before changing it.
* **I say this now** (on a framing you avoid): takes it off your list of framings you avoid.

When nothing is flagged you see **No stance conflicts**. After you edit the draft, **Re-check** runs it again.

{% hint style="info" %}
Stance never rewrites anything on its own, and a new claim is a question, not a failure. The result (On voice, Check one line, Off voice) comes from fixed rules applied to each line, not from a score the model makes up.
{% endhint %}

#### What Stance needs

* A trained voice, and a claim bank with something in it. With an empty bank there is nothing to check against yet, and Studio says so. See Claim bank and learning from edits.
* 1 credit per check. A check that fails is refunded.

#### On a phone

The strip under the draft shows your Style count and a Stance chip: "Stance: on voice", "Stance: check one line", "Stance: off voice", or "Stance: not checked".

#### In chat and from assistants

audenAI chat and the assistant tool `score_voice_match` give the same two results. See Ask audenAI and Connect VoiceMoat to an assistant.

#### Who can use it

|        | Free | Trial | Creator | Pro | Enterprise |
| ------ | ---- | ----- | ------- | --- | ---------- |
| Style  | no   | yes   | yes     | yes | yes        |
| Stance | no   | yes   | yes     | yes | yes        |

Style and Stance come with the trial and the paid plans. A Free plan's one free training builds the Profile DNA; the checks open when you start a trial or a plan.

#### Common questions

**Why is there no single voice score anymore?** A single number could look high while one line said something you would never say. Two results tell you what is off and where.

**Does Stance push opinions on me?** No. A new claim shows as "Check one line", a question for you, and nothing changes unless you pick a rewrite.

**Why does it say "A story you haven't told"?** The draft has a personal story or number that is not in your claim bank. audenAI only uses stories and numbers from your bank. If it is true, tap **This is me now** and it is added.

**Does Improve change my stance?** Not by default. Improve keeps your opinion, your opener, and your stories unless you tick **Also tune for reach**. See Tones and Improve.
