You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation using Clean Architecture principles.
Guides Flutter feature design through collaborative brainstorming and Clean Architecture principles.
/plugin marketplace add vp-k/flutter-craft/plugin install vp-k-flutter-craft-plugins-flutter-craft@vp-k/flutter-craftThis skill inherits all available tools. When active, it can use any tool Claude has access to.
Help turn ideas into fully formed Flutter designs and specs through natural collaborative dialogue.
Start by understanding the current project context (pubspec.yaml, lib/ structure, existing features), then ask questions one at a time to refine the idea. Once you understand what you're building, present the design in small sections (200-300 words), checking after each section whether it looks right so far.
Announce at start: "I'm using the flutter-brainstorming skill to design this feature."
Check project context first:
# Check pubspec.yaml for dependencies
cat pubspec.yaml
# Check existing lib/ structure
ls -la lib/
# Check existing features
ls -la lib/features/ 2>/dev/null || echo "No features folder yet"
Ask questions one at a time:
Key questions to explore:
Propose 2-3 different approaches with trade-offs:
Example for state management:
Approach A: Riverpod + Freezed (Recommended for new projects)
- Pros: Compile-time safety, immutable states, no boilerplate with codegen
- Cons: Learning curve, requires build_runner
- Best for: Most Flutter projects
Approach B: BLoC Pattern
- Pros: Separation of concerns, testable, scalable
- Cons: More boilerplate
- Best for: Complex state, multiple streams, large teams
Approach C: Provider + ChangeNotifier
- Pros: Simple, familiar
- Cons: Less structured, harder to test
- Best for: Simple local state, small projects
Riverpod + Freezed Pattern Example:
// State with freezed
@freezed
class AuthState with _$AuthState {
const factory AuthState.initial() = _Initial;
const factory AuthState.loading() = _Loading;
const factory AuthState.authenticated(User user) = _Authenticated;
const factory AuthState.error(String message) = _Error;
}
// Notifier with riverpod_generator
@riverpod
class Auth extends _$Auth {
@override
AuthState build() => const AuthState.initial();
Future<void> login(String email, String password) async {
state = const AuthState.loading();
try {
final user = await ref.read(authRepositoryProvider).login(email, password);
state = AuthState.authenticated(user);
} catch (e) {
state = AuthState.error(e.toString());
}
}
}
Lead with your recommendation and explain why.
Present the design in sections of 200-300 words:
Feature Overview
Clean Architecture Structure
lib/features/<feature_name>/
├── domain/
│ ├── entities/ # Business objects
│ ├── repositories/ # Repository interfaces
│ └── usecases/ # Business logic
├── data/
│ ├── models/ # DTOs (fromJson, toJson)
│ ├── datasources/ # Remote & Local data sources
│ └── repositories/ # Repository implementations
└── presentation/
├── bloc/ or provider/ # State management
├── widgets/ # Reusable UI components
└── screens/ # Full screen widgets
Data Flow
UI/UX Design
Testing Strategy (priority-based)
Ask after each section: "Does this look right so far?"
Write the validated design to:
docs/plans/YYYY-MM-DD-<feature>-design.md
Design document template:
# <Feature Name> Design
## Overview
[1-2 sentences describing the feature]
## User Stories
- As a user, I want to...
## Clean Architecture
### Domain Layer
- Entities: [list]
- UseCases: [list]
- Repository interfaces: [list]
### Data Layer
- Models: [list]
- DataSources: [list]
- Repository implementations: [list]
### Presentation Layer
- State Management: [BLoC/Provider/Riverpod]
- Screens: [list]
- Widgets: [list]
## Data Flow
[Diagram or description]
## API Contract
[If applicable]
## Testing Plan
- Unit tests: [list]
- Widget tests: [list]
- Integration tests: [if needed]
## Dependencies
[New packages needed in pubspec.yaml]
Commit the design document to git:
git add docs/plans/
git commit -m "docs: add <feature> design document"
Ask: "Ready to create an implementation plan?"
If yes:
flutter-craft:flutter-planning to create detailed implementation planflutter-craft:flutter-worktrees if isolated workspace is neededAfter completing brainstorming and documenting the design, you MUST invoke: → Skill tool: flutter-craft:flutter-planning
This is NOT optional. The workflow is incomplete without a detailed implementation plan.
Never skip brainstorming because:
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.