ScriptGrain makes Meta Muse better at being you

By Jack Stovell X · Instagram · LinkedIn · published 2026-09-20 · updated 2026-09-30

Meta Muse can write for you. It cannot measure you. ScriptGrain measures your voice as 45 attributes from writing you have already published, connects to Meta Muse, and scores every draft against the measurement, so the assistant you already use writes as you rather than as its average.

Start with the prompt builder below: paste your writing and get a system prompt written in measured numbers, ready for custom instructions or a project. The connector does the same with the full profile.

How to connect

Works with Meta Muse (meta.ai, US).

  1. Measure your voice. Create a free ScriptGrain account and paste three or more pieces you wrote (3,000 words works best). The profile usually takes under a minute and costs nothing.
  2. Connect. Ask Muse to build a custom integration to ScriptGrain with the MCP server URL `https://mcp.scriptgrain.com/mcp`. Muse asks for credentials: mint an API key at Settings, then API, and enter it in Muse's secure credential store. Once ScriptGrain is listed in the Muse connector directory, connecting is a sign-in on the consent page instead.
  3. Write, then score. Muse holds the connection across conversations, so "write this in my voice" works without re-explaining who you are, and "is this on voice?" scores any text you paste.

What changes

Without a profile, Meta Muse writes from a description: whatever you can say about your own style. Most of a voice is what you cannot say: contraction rate, clause ordering, how often you open with a question, the words you reach for without noticing. With ScriptGrain connected, Meta Muse writes from the measured profile and every draft comes back with a voice-match score and the exact features that moved. On-voice drafts score 0.85 to 0.95; a clearly different voice scores under 0.45. Paste any two pieces into the free off-voice check to see the same measurement with no account.

What stays free

The profile, the connector, scoring any text against your profile, and comparing pieces with no profile at all are free on every plan. Generating drafts in your voice, polishing toward a target and the monthly voice monitor are paid.

What Muse reads from the profile

Muse doesn't get a paragraph about you. It gets numbers. The profile is 45 attributes in 8 layers, and Muse reads them live from ScriptGrain each time it drafts. No pasted prompt, nothing to go stale.

So what does that look like in practice? Average sentence length and its variance. Contraction rate. Comma density, plus your semicolon and dash habits. Formality on a 0 to 10 scale. Confidence against hedging. How you open (question, claim, scene) and how you order an argument: evidence-first, conclusion-first or narrative. Even the words your paragraphs tend to start with.

To be fair, Muse is a capable drafter, and it will follow a style description. The gap is the description itself. You can tell it "conversational but sharp" and it will try. You can't easily tell it "a contraction rate around half, sentence lengths that swing wildly, conclusions before evidence." Nobody notices those things about their own writing. The profile states them for you, in figures Muse can work to.

One more tool sits in the connection: `get_profile`, which lets you ask Muse what it currently knows. Worth a look before your first draft.

A drafting session, step by step

Say you write a dry, direct newsletter and need a piece on a product launch. Here's a realistic run.

  1. Brief it plainly. Give Muse the topic, the reader and the point you want made. Something like: "Write 400 words for subscribers on why we delayed the launch. Use my voice profile." You don't describe your style. That's the whole idea.
  2. Let it draft. Muse pulls the profile, writes against it, and generates the draft. On a paid plan that costs 1 credit per draft.
  3. Read the score. The draft returns with a voice match between 0 and 1. On-voice drafts typically land 0.85 to 0.95. Partial matches sit between 0.5 and 0.8.
  4. Check the deltas. Each feature shows the draft's value against yours: sentence length, hedging, contractions. This is the useful bit. A 0.78 tells you something is off; the deltas tell you what.
  5. Edit and save. Fix the lines that still feel wrong, then use `save_edit`. It's free, and the app learns from the edits you keep.

And if you only want a check, skip drafting entirely. Paste text you wrote (or text Muse wrote from a plain prompt) and ask "is this on voice?" Scoring is free.

When a draft scores off voice

So the score comes back at 0.6. Don't rewrite from scratch. Read the deltas first.

Usually one or two features carry most of the gap. Say the draft's average sentence length is nearly double yours, and its hedging is well above your profile value. That's a specific, fixable problem: cut the long sentences, strip the "perhaps" and "arguably" padding, state the claim plainly. You've just turned a vague sense of wrongness into a two-line edit.

Next, try polish. It revises a draft toward a target score, 0.9 by default, so you can nudge a 0.72 upward without hand-editing every line. Then read the result anyway. A score measures the surface of your voice; you still judge whether the point lands.

Two further checks, in order:

A score under 0.45 means a clearly different voice. At that point, check you pointed Muse at the right profile. It happens.

Questions

Do I need a paid plan to connect ScriptGrain to Meta Muse?

No. The connector, the profile and voice-match scoring are on the Free plan. Paid plans add generation credits, polishing and the monthly voice monitor.

Does Meta Muse send my conversations to ScriptGrain?

No. Meta Muse calls ScriptGrain only when it uses a tool (score this text, generate this brief), and ScriptGrain receives only what that call contains. Your API key or sign-in token never appears in the conversation.

What is the voice-match score measuring?

Stylistic alignment: sentence rhythm, punctuation habits, pronoun mix, formality, argument structure and the rest of the measured attributes, compared feature by feature against your profile. It says whether the draft sounds like you, not whether it is any good.

Related