CLAUDE LABJP
2.1.274 — This round is aimed at people running Claude Code unattended: a warning when memory runs critical, an environment variable that bounds how long the first turn waits for MCP servers, and an end to sessions retrying a 400 forever10/07 — The old management-settings spelling is accepted until noon PT on October 7, nineteen days from now. Deprecation warnings have been showing since September 10WINUPD — Reports keep coming in of a Windows update leaving Cowork unable to mount a single host folder. Removing the update is still the only workaround anyone has foundNEW — Stopping before you hit the limit: a record of rebuilding the day around the five-hour windowBING — Seven in ten of the people who actually read these pages arrive from Bing. Search has more than one front doorEXCEL — Before handing over a spreadsheet, decide which columns it may read and which it may not2.1.274 — This round is aimed at people running Claude Code unattended: a warning when memory runs critical, an environment variable that bounds how long the first turn waits for MCP servers, and an end to sessions retrying a 400 forever10/07 — The old management-settings spelling is accepted until noon PT on October 7, nineteen days from now. Deprecation warnings have been showing since September 10WINUPD — Reports keep coming in of a Windows update leaving Cowork unable to mount a single host folder. Removing the update is still the only workaround anyone has foundNEW — Stopping before you hit the limit: a record of rebuilding the day around the five-hour windowBING — Seven in ten of the people who actually read these pages arrive from Bing. Search has more than one front doorEXCEL — Before handing over a spreadsheet, decide which columns it may read and which it may not
Articles/Claude.ai
Claude.ai/2026-03-21Beginner

Claude Custom Styles — How to Tailor AI Responses to Your Preferences

A deep dive into Claude's Custom Styles feature. Learn how to use preset styles, create your own, and configure styles for business, learning, and creative use cases to dramatically improve your AI experience.

Claude48Custom Stylespersonalizationprompting3settings9tone2writing3

What Are Custom Styles?

Claude includes a Custom Styles feature that lets you adjust the tone, format, and overall feel of its responses. While the default Claude provides well-balanced, general-purpose answers, you might prefer something more concise for work, more thorough for learning, or more creative for brainstorming. Custom Styles let you set those preferences once, so Claude consistently responds the way you want — without having to repeat instructions in every prompt.


Getting Started with Preset Styles

Claude comes with several built-in preset styles that you can switch between from the style selector near the chat input area.

Normal

The default response style. It strikes a balance between thoroughness and brevity, and works well across a wide range of questions. Stick with this if you don't have a strong preference or if you're new to Claude.

Concise

A stripped-down, just-the-essentials style. When you need quick answers without lengthy explanations, this is your go-to. It's especially useful during busy work sessions or when you're firing off a series of short questions.

Explanatory

A style that takes its time to walk you through concepts step by step. Great for learning new technologies, understanding complex topics, or getting answers that include background context and worked examples.

Formal

A polished, professional tone suited for business writing, academic contexts, and official communications. Use this when drafting reports, reviewing email copy, or discussing research topics where precision matters.


Creating Your Own Style

If the presets don't quite fit your needs, you can build a fully custom style from scratch.

How to Create One

Open Settings and navigate to the "Styles" section. Click "Create Custom Style" to open the editor, where you describe your preferred response style in plain text.

For example, you might write: "Keep responses under 200 words unless I ask for detail. Use bullet points for lists. When discussing code, always include a working example with inline comments. Avoid hedging language — be direct and confident."

The more specific your instructions, the better Claude can match your expectations.

Generating a Style from Sample Text

Another approach is to paste in a sample of writing you admire — whether it's your own blog post, a technical document, or a newsletter you enjoy reading. Claude can analyze the tone, structure, and vocabulary of that sample and generate a style profile that mirrors it.


Recommended Styles by Use Case

Business and Professional Work

For business contexts, a style that leads with conclusions and backs them up with concise evidence tends to work best. Try something like: "Answer with the conclusion first, followed by 2-3 supporting points. Use numbers and specifics over vague language. Keep total response length under 150 words unless the topic demands more."

Programming and Technical Learning

If you're learning to code, configure a style that pairs executable examples with clear explanations. A setting like: "Always provide working code examples. Add comments explaining each significant line. After the code block, explain why this approach was chosen and mention common pitfalls." This turns Claude into a patient tutor who shows rather than just tells.

Creative Writing

For creative projects, you want Claude to think expansively rather than converge on a single answer. Consider: "Offer at least three distinct options or directions. Include at least one unexpected or unconventional suggestion. When critiquing my work, always pair criticism with a concrete alternative."

Academic Research

Research work demands accuracy and intellectual honesty. A style like: "Cite the basis for every claim. Flag information you're uncertain about. Define technical terms on first use. Distinguish between established consensus and emerging perspectives." helps maintain scholarly rigor in your conversations.


Switching Between Styles

You can create multiple custom styles and switch between them depending on your task. For instance, you might maintain a "Work" style, a "Learning" style, and a "Casual" style, then pick the right one at the start of each conversation.

Styles apply to the entire conversation once selected, but you can also change styles mid-conversation. This opens up interesting workflows — start with an explanatory style while you're wrapping your head around a new concept, then switch to concise mode once you're ready to move fast.


Combining Styles with Projects

Custom Styles become even more powerful when paired with Claude's Projects feature.

Projects let you upload reference documents and set project-specific System Instructions. While Custom Styles control how Claude answers (tone, format, length), System Instructions control what context Claude brings to those answers (domain knowledge, technical constraints, team conventions).

For example, you could set up a project with your API documentation and a System Instruction like "Assume all code examples target Python 3.12 and use the httpx library." Then apply your "Concise" custom style. Inside that project, you'll get tight, focused Python code answers grounded in your actual API specs.


Tips for Writing Effective Styles

A few principles will help you get the most out of custom styles.

First, be specific rather than vague. "Write clearly" doesn't give Claude much to work with. "Use short sentences. Avoid jargon. If a technical term is necessary, define it in parentheses." gives Claude a concrete target.

Second, include both positive and negative instructions. Don't just say what you want — also say what you don't want. "Do not use bullet points for explanations" or "Never start a response with 'Great question!'" can be just as helpful as affirmative instructions.

Third, revisit and refine your styles periodically. Your needs evolve as your skills grow and your projects change. A style that worked perfectly three months ago might need tuning to match where you are today.


Wrapping Up

Claude's Custom Styles feature transforms a general-purpose AI into a personalized assistant that communicates exactly the way you prefer. Whether you need crisp professional responses, detailed educational walkthroughs, or freewheeling creative brainstorming, there's a style configuration that fits.

If you haven't explored Custom Styles yet, start by toggling between "Concise" and "Explanatory" on your next few conversations. Once you see how much the experience changes, you'll want to craft your own.

Share

Thank You for Reading

Claude Lab is ad-free, supported entirely by members like you. We publish practical guides daily with implementation code, benchmarks, and production-ready patterns. If you've found it useful, we'd love to have you on board.

  • Copy-paste ready implementation code
  • New advanced guides published daily
  • $5/mo or $15 for lifetime access
View Membership →

If you found this article helpful, a small tip ($1.50) would mean a lot to us. Your support helps keep this site ad-free and covers server and hosting costs.

Related Articles

Claude.ai2026-09-19
When a proofreading request comes back as a rewrite: keeping the original wording
Ask for proofreading and you often get a polished replacement instead. Fixing the output shape first, splitting edits into three layers, and showing one keep-and-change example pair will hold on to the author's sentence endings and word order.
Claude.ai2026-09-15
Where to Put an Instruction That Keeps Coming Back: Chat, Style, Settings File, or Hook
No comments, please. Do not translate the product name. When an instruction drifts back a few turns later, the fix is usually the location, not the wording. I compared the same sentence across four layers and settled on a rule for which instruction belongs where.
Claude.ai2026-09-14
Picking Up Where You Left Off: When to Lean on Claude's Memory and When to Ask It to Search
Claude's memory and past-chat search are two different mechanisms. How to tell which one answered you, why project boundaries block searches, why deleting a chat leaves its memory behind, and which plans get what.
📚RECOMMENDED BOOKS
Build a Large Language Model (From Scratch)
Sebastian Raschka
LLM Dev
Prompt Engineering for LLMs
Berryman & Ziegler
Prompting
AI Engineering
Chip Huyen
AI Eng
* Contains affiliate links