Brainstorming — Design Before Code
> **Type:** Rigid process (follow exactly)
Best use case
Brainstorming — Design Before Code is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
> **Type:** Rigid process (follow exactly)
Teams using Brainstorming — Design Before Code should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/brainstorming/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How Brainstorming — Design Before Code Compares
| Feature / Agent | Brainstorming — Design Before Code | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
> **Type:** Rigid process (follow exactly)
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Brainstorming — Design Before Code > **Type:** Rigid process (follow exactly) > **Trigger:** Any creative work — creating features, building components, adding functionality > **Hard Gate:** NO implementation until design is approved ## Iron Law ``` NO IMPLEMENTATION UNTIL DESIGN IS APPROVED ``` ## Purpose Turns vague ideas into fully formed design specifications through structured, collaborative dialogue. Prevents the "jump straight to code" anti-pattern that leads to wasted effort and architectural mistakes. ## The Checklist (7 Steps) ### Step 1: Explore Project Context - Check existing files, docs, README, recent commits - Understand the current architecture before proposing changes - Look for related existing functionality ### Step 2: Offer Visual Companion (if applicable) - If the topic involves visual/UI questions, offer a visual mockup - This MUST be its own message (not buried in other content) - Use diagrams, wireframes, or component sketches as appropriate ### Step 3: Ask Clarifying Questions - One question at a time — never overwhelm - Prefer multiple choice over open-ended - Ask about: scope, constraints, edge cases, integration points ### Step 4: Propose 2-3 Approaches - Each with clear trade-offs - Include a recommendation with rationale - Consider: complexity, maintainability, performance, existing patterns ### Step 5: Present Design - Scale sections to complexity (few sentences if simple, detailed if nuanced) - Get approval per section for complex designs - Include: data flow, component boundaries, error handling, testing strategy ### Step 6: Write Design Document - Save to `docs/specs/YYYY-MM-DD-<topic>-design.md` - Include all decisions, rationale, and approved approaches - This becomes the source of truth for implementation ### Step 7: Transition to Planning - Invoke `writing-plans` skill (and ONLY writing-plans) - Pass the approved design document as input - Do NOT start implementing — plans come first ## Key Principles - **One question at a time** — never overwhelm the user - **YAGNI ruthlessly** — remove unnecessary features during design - **Scale to complexity** — simple features get simple designs - **Design for isolation** — break into units with clear boundaries - **User instructions override** — if user says "skip design", respect that ## Rationalization Table | Excuse | Reality | |--------|---------| | "This is too simple to design" | Simple features still need scope definition. 5 minutes of design saves hours of rework. | | "I already know how to build this" | Your knowledge isn't the issue. Design prevents scope creep and missed requirements. | | "The user wants it fast" | Fast and wrong is slower than design-first and right. | | "I can design as I code" | That's called hacking. Design THEN code. | | "The requirements are clear" | Clear to whom? Verify with clarifying questions. | ## Red Flags - Starting to write code before Step 5 is approved - Skipping clarifying questions because "it's obvious" - Proposing only one approach (always offer alternatives) - Design document missing error handling or testing strategy - Moving to implementation without writing the design doc
Related Skills
web-design-guidelines
Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".
Verification Before Completion — The Honesty Enforcer
> **Type:** Rigid (follow exactly)
ux-interaction-design
UX interaction design patterns including micro-interactions, UI state machines, transition specifications, form UX, animation principles, and feedback patterns. Use when designing interactions, defining component states, or specifying animations.
ui-visual-design
Visual design mastery including color theory, typography systems, visual hierarchy, hero section patterns, page section patterns, design tokens, and modern design trends. Use when creating visual designs, building design systems, designing hero sections, or establishing design tokens.
ui-ux-designer
UI/UX 设计师专家,精通界面设计、交互设计、用户体验和设计系统
mobile-design
Mobile-first design and engineering doctrine for iOS and Android apps. Covers touch interaction, performance, platform conventions, offline behavior, and mobile-specific decision-making. Teaches principles and constraints, not fixed layouts. Use for React Native, Flutter, or native mobile apps.
frontend-design
Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.
design-review
Design critique and review methodology including heuristic evaluation, consistency checking, accessibility compliance, and visual quality assessment. Use when reviewing design artifacts, evaluating UI quality, or auditing design compliance.
design-handoff
Design handoff protocols including artifact format standards, lifecycle state management, inter-agent communication, and quality gate procedures. Use when managing design lifecycle state, coordinating between agents, or formatting handoff artifacts.
YAML Prompt Library
> Store reusable AI prompts as YAML files with structured messages, variables, and test data for version-controlled prompt engineering.
writing-skills
Use when creating new skills, editing existing skills, or verifying skills work before deployment
Writing Plans — TDD-Sized Task Breakdown
> **Type:** Rigid process (follow structure exactly)