Give a Claude Connector Your Writing Style: Setup Guide

TL;DR: Claude custom connectors let Claude call an outside service mid-conversation. Point one at a Style Profile built from your own writing and Claude stops guessing at your voice: it loads the profile, rewrites drafts in it, and checks a draft against your measured sentence length, comma rate and vocabulary variety. Setup is Settings > Connectors > Add custom connector with the MyWritingTwin OAuth URL. The same profile then serves Claude Code, Claude Desktop and any MCP agent you build.

You wired a Claude connector into your workflow, ran it on a real task, and the first draft came back reading like a press release. Correct, polite, and nothing like you.

That gap is not a Claude problem. Claude has no idea how you write, because nothing about you is in the model, and a connector is the cleanest way to change that.

What Does a Claude Connector Actually Change?

A connector gives Claude a live line to an outside service, and a writing-style connector gives it a live line to a profile of how you write. Instead of pasting a description of your voice into every chat, Claude fetches the real thing when it needs it. The profile stays in one place, and every surface that uses the connector reads the same source.

Anthropic calls these custom connectors. They are remote MCP (Model Context Protocol) servers you add under Settings > Connectors, and per Anthropic's connector documentation they are available on the Free, Pro, Max, Team and Enterprise plans, on Claude web, Claude Desktop and Cowork. Once a connector is attached, Claude can call its tools in the middle of a conversation, the same way it searches the web or reads a file.

The MyWritingTwin connector exposes two kinds of things:

• A resource, mwt://profiles/current, holding your active Style Profile in system-prompt form. Clients that discover context through MCP resources can read it without a tool call.
• Tools to rewrite a draft in your voice, check a draft against your measured patterns, export the profile in several formats, and deposit new writing samples into your account.

That second list is what makes this different from a settings page. The same connector that serves a claude.ai conversation also serves Claude Code, Claude Desktop, and any agent that speaks MCP. One profile, many consumers. If you build agents on Claude, that is the property you were looking for.

Why Doesn't Claude Already Know Your Writing Style?

Because nothing about you is in the model. Claude is trained on public text and tuned to be useful to everyone, which makes its default voice the average of everyone. Model capability and user-data conditioning are separate layers: Anthropic keeps improving the first, and no lab can supply the second. Your emails are not in the training data, and a general-purpose model is not tuned to prefer them even if they were.

That is why the usual fix, telling Claude to "write in a professional yet friendly tone", changes so little. "Professional yet friendly" describes thousands of people. A voice lives in sentence length, qualifier placement, transitions and punctuation. Claude does not know your tone; it knows words that describe tone, and those words map to the same generic register for every user who types them.

A short comparison of what a label carries versus what a profile records:

The right-hand column is what a Style Profile stores, and it is what the connector hands Claude. We covered the extraction itself in how AI style extraction works; the short version is that the profile is evidence of how you write, not a hope about how you might sound.

One more point, because it decides whether this is worth doing at all. A better model does not make a personal profile less necessary. It makes the profile more valuable. The better Claude gets at following nuanced instructions, the more faithfully it executes a precise profile. Every model generation narrows the execution gap and leaves the specification gap exactly where it was, since the specification has to come from your writing. When we profiled five AI models like human writers, each had a measurable house style. None of them matched a particular person, and none of them will, because that data is yours.

How Do You Connect Claude to Your Writing Style?

Three steps: build the profile, add the connector, ask Claude to use it. Budget fifteen minutes if you already have writing samples on hand, most of it waiting for generation.

Step 1: Build a Style Profile from your own writing

Create a free account and add writing you actually sent: emails, posts, reports, messages. Not AI drafts you edited, and not the writing you wish you produced. The profile models whoever wrote the text, so the samples have to be yours. The first use case is free with no card, and it is built from roughly 2,000 words of samples.

Generation takes 30 to 90 seconds. You get a Writing DNA analysis plus deployable exports: a Master Prompt, a Runtime Block in JSON for agents, and the system-prompt format the connector serves. If you have not seen what a profile contains, start with what a Writing Style Profile is.

Step 2: Add the connector in claude.ai

1. Open claude.ai and go to Settings > Connectors.
2. Choose Add custom connector at the bottom of the connector list.
3. Name it MyWritingTwin and enter the remote MCP server URL: https://www.mywritingtwin.com/api/auth/mcp
4. Click Add, then Connect. Claude opens a MyWritingTwin sign-in screen (OAuth 2.1). Approve it and the connection is live.

Use the address ending in /api/auth/mcp, not the plain /api/mcp one. The plain endpoint is for bearer-key clients and never prompts for a login, so claude.ai will not show a Connect step there. On a Team or Enterprise plan, if the Add custom connector option is missing, your admin controls it.

Step 3: Ask Claude to use it

Turn the connector on in the conversation's tools menu, then ask in plain language:

• "Load my writing profile and rewrite this in my voice: [paste draft]"
• "Check this draft against my profile and tell me which constraints fail."
• "Draft a reply to this thread in my client-email mode."

Claude picks the right tool from the request. You do not need to name the tools, though it helps to know what they do.

Claude Code and Claude Desktop

Same server, different auth. Create an API key on your dashboard and register the server:

claude mcp add --transport http mywritingtwin https://www.mywritingtwin.com/api/mcp -h "Authorization: Bearer mwt_sk_..."

For Claude Desktop, add the same URL and header under mcpServers in claude_desktop_config.json. The full reference, including the REST endpoints behind the tools, is in the API documentation.

Which Tools Does the Connector Give Claude?

Six tools carry most of the work. The table lists what each one does and when Claude reaches for it.

The one worth understanding is mwt_check_draft. Claude's own opinion on whether a draft matches your voice is not trustworthy. We measured assistants inventing passing numbers, then drifting further from target on each revision, so the check runs as real code on our server and returns the actual numbers. If a constraint fails, Claude revises and checks again, and stops when everything passes. A self-report cannot be faked into that loop.

Modes are the other feature to know. A profile is a matrix of language, audience and medium, and the connector can serve a single cell of it. Your client emails and your team Slack messages are different registers, and Claude should get the one you asked for, not a blend.

How Is This Different From Claude Projects or Custom Styles?

Projects and Custom Styles are containers. The connector is a source. Both are worth using, and they do different jobs.

Our guide to making Claude sound like you walks through Projects and system prompts in detail, and that setup still works. The limit is what those containers hold. A Project holds a description you wrote yourself, with the blind spots self-description always has. You describe the writer you want to be, and you cannot see the patterns you produce on autopilot.

There is also a structural difference. A Project is a session; a profile is an asset. A description typed into Claude stays in Claude. A profile you own can be exported as a Runtime Block and dropped into ChatGPT, Gemini, a Cursor rule or your own agent, which is the point of deploying one Writing Twin across every AI platform. The connector does not replace your Project. Paste the profile export into the Project instructions if you like, and keep the connector for what static text cannot do: check and rewrite against measured constraints, and stay current when you regenerate.

What Should You Do When the Output Still Drifts?

Run the check instead of rereading the draft. Most drift has one of four causes, and each has a fix that takes a minute.

1. Wrong mode. Ask for the register you need by name: "use my client-email mode". A blended profile reads flatter than any single cell.
2. Claude eyeballed it. Ask it to run mwt_check_draft and show the table. A failing constraint is a specific instruction; "sounds a bit off" is not.
3. Thin samples. If the failing constraints cluster in one medium, deposit more of that medium with mwt_import_samples and regenerate. The profile is only as wide as the writing behind it.
4. The connector is off. Custom connectors are enabled per conversation in the tools menu. If Claude answers from memory, check the toggle.

Then the honest limit. A profile constrains rhythm, vocabulary, punctuation and structure. It carries no opinions and no facts. Claude will write the paragraph the way you write paragraphs, and you still have to tell it what the paragraph says. Anyone who promises that a style layer makes the content yours as well is overselling.

What to Do Next

1. Build your free Style Profile from writing you actually sent.
2. Add the custom connector in claude.ai with the OAuth URL above and confirm the Connect step appears.
3. Run one real draft through mwt_rewrite_in_voice, then mwt_check_draft, and read the numbers.
4. If you build on Claude Code or your own agents, register the same server with an API key and stop maintaining a voice prompt in every repo.

Connectors are the first Claude surface where your writing style can live outside the chat and be read on demand. Your Writing Twin for Claude, ChatGPT, Gemini and any MCP agent is one profile, built once, and the connector is how Claude reads it. The Claude platform page has the shorter version if you want to send it to a colleague.