Design Review logo

Design Review

Organization
bitwarden
design-review

Bitwarden design team's Code of Conduct combined with the 30/60/90 critique framework — stage-appropriate critique, product-not-designer focus, content evaluated alongside visual design at 60% and 90%.

Overview

Publisherbitwarden
Repositoryai-plugins
Skill namedesign-review
Stars
149
Forks
19
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by bitwarden on GitHub. Read the source before you install it.

Installation

Install the Design Review AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/bitwarden/ai-plugins.git /tmp/ai-plugins
mkdir -p .claude/skills
cp -r /tmp/ai-plugins/plugins/bitwarden-designer/skills/design-review .claude/skills/design-review
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Design Review in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Design Review on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Design Review is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Design Principles & Critique

This skill grounds design feedback in the Bitwarden design team's Code of Conduct and the 30-60-90 maturity framework. The Code of Conduct and 30/60/90 framework originate from the designer-agent-skills branch of bitwarden/clients (authored by the design team) and have no separate Confluence canonical — this skill is the authoritative source within Claude tooling. Critique the product, not the designer. Feedback must be actionable, stage-appropriate, and rooted in product goals — not personal preference.

Cross-plugin dependency. This skill composes content-style-guide (at 60% and 90% stages) and using-figma (when the design lives in a Figma file). Both ship in the bitwarden-design-tools plugin, which is required alongside bitwarden-designer — install both for the full composition to work.

Step 1: Identify the stage

Before giving feedback, identify (or ask) what stage the design is at. The kind of feedback that's useful depends entirely on this.

  • 30% — a rough idea. The designer is exploring direction. Easy to pivot or throw away. Looking for: ideas and impressions, whether this is something we should do, whether it's the right direction, how to move the concept forward, go/no-go on the idea.
  • 60% — a first draft of a set concept. Significant time has gone into this; direction shouldn't change drastically without strong reason. Looking for: whether 30% critique was addressed, visual/graphic feedback, feedback on interactive components, ways to expand the concept.
  • 90% — last check before development. Should already be tested with real users. No drastic changes expected. Looking for: whether 60% critique was addressed, nitty-gritty grammar, finalizing copy, final check on the minutiae.

If the user doesn't say the stage, ask. Don't give 90%-style nitpicks on a 30% sketch, and don't suggest sweeping direction changes on a 90% design.

Step 2: Critique the product, not the designer

A good design meets its goals. A bad design does not meet its goals. Likes and dislikes are irrelevant.

  • Talk about strengths, not just weaknesses. Good critique empowers — understanding what works helps decide what to keep.
  • Separate like/hate from good/bad. Consider product goals over personal opinions.
  • Ask questions instead of making assumptions. If something isn't clear, ask why a decision was made.
  • Don't try to design a better solution on the spot. Focus on what about the current design isn't meeting its intended purpose. The designer can address it when they have more time.

Phrasing

Instead ofTry
"Why did you do that?""What are you trying to achieve by doing x?"
"I don't like it""I'm not sure that x makes it clear to users they can y"
"Why don't we just…"(skip — don't design on the spot; describe the gap instead)

Step 3: Filter feedback through the Code of Conduct

The Bitwarden design team operates by five principles. Let them shape what to flag and how.

  1. We design proactively, not reactively. Lead with strategy, planning to innovate and shape the product's future rather than just reacting to demand. Create space for intentionality, so every designer can be thoughtful, detail-oriented, and proud of their work. Flag missed opportunities for intentional, forward-looking design — especially at the 30% stage.

  2. We design with empathy, verified by insights. User voices guide decisions through regular conversations, research, and data. Bring customers to the forefront of every discussion, empowering partners in Product and Engineering to think customer-first. Every feature shipped should be user-centered, tested, and proven. When a design choice is unsupported by insight, surface it as an open question — not a flaw, but a gap worth closing before 60%.

  3. We design with confidence and humility. Trust expertise while remaining open to being wrong. Navigate ambiguity together, make clear decisions, and move forward — staying open to changing course as more is learned. Even minor iterations can transform outdated experiences into something to be truly proud of. Frame feedback as contribution to a shared solution, not as a verdict.

  4. We work as a unified team. Collaboration with Product and Engineering should be smooth and transparent. Design files should be clear, easy to navigate, and well-organized with shared understanding. Operate with clarity, confidence, and care — trust is the foundation, and everyone is encouraged to contribute their best. Flag ambiguity that will make handoff painful.

  5. A year from now, we want a Bitwarden UI we're proud to put our names on — one that earns the trust of millions because we designed it with the trust of each other. Hold feedback to that bar.

Step 4: Evaluate content alongside visual design

At 60% and especially 90%, evaluate user-visible copy (button labels, headings, error messages, empty states, helper text, etc.) against the content-style-guide skill — voice, tone, sentence case, no ampersands, meaningful link text, gender-neutral pronouns, no spatial language, and the rest. Treat content findings as first-class critique points, not afterthoughts.

Skip content nitpicks at 30% — direction, not copy, is the question at that stage.

Composing with other skills

  • content-style-guide. Compose at 60% and 90% stages to evaluate user-visible copy alongside visual design — voice and tone, sentence case, no ampersands, accessibility rules. Skip at 30%, where copy is too early to critique.
  • using-figma. When the design under review lives in a Figma file, compose to read the context. Start with get_screenshot + get_metadata to orient; pull get_variable_defs when tokens are part of the critique (off-system colors, inconsistent spacing).

Output format

Structure the critique as:

  1. Stage — confirm or ask (30% / 60% / 90%).
  2. Strengths — what's working and should stay.
  3. Questions — what isn't clear; what to ask the designer to clarify.
  4. Actionable feedback — specific, product-goal-anchored observations the designer can address. Match the granularity to the stage. At 60%/90%, include content observations tied to the content-style-guide.

Keep feedback specific. "The CTA hierarchy makes it unclear which action is primary" beats "the buttons feel off." Tie each point back to a user or product goal where you can.

Frequently asked questions

What does the Design Review AI skill do?

Bitwarden design team's Code of Conduct combined with the 30/60/90 critique framework — stage-appropriate critique, product-not-designer focus, content evaluated alongside visual design at 60% and 90%.

Why use Design Review on TypingMind?

Because you install it once and use it with any model. Design Review is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Design Review in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/bitwarden/ai-plugins/tree/main/plugins/bitwarden-designer/skills/design-review. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Design Review?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Design Review?

As many as you like. As long as a model supports skills, you can use Design Review with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Design Review AI skill free?

It is published on GitHub by bitwarden. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇