Brainstorming Ideas Into Designs
Help turn ideas into fully formed designs and specs through natural collaborative dialogue.
Start by understanding the current project context, then ask only questions whose
answers materially affect the design. Present a design and obtain approval before
implementing substantial or ambiguous changes. Skip this workflow for small,
well-specified fixes and routine configuration changes.
Checklist
You MUST create a task for each of these items and complete them in order:
- Explore project context — check files, docs, recent commits
- Ask clarifying questions — one at a time, understand purpose/constraints/success criteria
- Propose 2-3 approaches — with trade-offs and your recommendation
- Present design — in sections scaled to their complexity and obtain approval
- Write a design document when useful — use the repository's preferred location
- Spec self-review — check placeholders, contradictions, ambiguity, and scope
- Transition to implementation — create a plan appropriate to the task
Process Flow
digraph brainstorming {
"Explore project context" [shape=box];
"Ask clarifying questions" [shape=box];
"Propose 2-3 approaches" [shape=box];
"Present design sections" [shape=box];
"User approves design?" [shape=diamond];
"Write design doc" [shape=box];
"Spec self-review\n(fix inline)" [shape=box];
"User reviews spec?" [shape=diamond];
"Invoke writing-plans skill" [shape=doublecircle];
"Explore project context" -> "Ask clarifying questions";
"Ask clarifying questions" -> "Propose 2-3 approaches";
"Propose 2-3 approaches" -> "Present design sections";
"Present design sections" -> "User approves design?";
"User approves design?" -> "Present design sections" [label="no, revise"];
"User approves design?" -> "Write design doc" [label="yes"];
"Write design doc" -> "Spec self-review\n(fix inline)";
"Spec self-review\n(fix inline)" -> "User reviews spec?";
"User reviews spec?" -> "Write design doc" [label="changes requested"];
"User reviews spec?" -> "Invoke writing-plans skill" [label="approved"];
}
The terminal state is an approved design with a concrete implementation plan.
The Process
Understanding the idea:
- Check out the current project state first (files, docs, recent commits)
- Before asking detailed questions, assess scope: if the request describes multiple independent subsystems (e.g., "build a platform with chat, file storage, billing, and analytics"), flag this immediately. Don't spend questions refining details of a project that needs to be decomposed first.
- If the project is too large for a single spec, help the user decompose into sub-projects: what are the independent pieces, how do they relate, what order should they be built? Then brainstorm the first sub-project through the normal design flow. Each sub-project gets its own spec → plan → implementation cycle.
- For appropriately-scoped projects, ask questions one at a time to refine the idea
- Prefer multiple choice questions when possible, but open-ended is fine too
- Only one question per message - if a topic needs more exploration, break it into multiple questions
- Focus on understanding: purpose, constraints, success criteria
Exploring approaches:
- Propose 2-3 different approaches with trade-offs
- Present options conversationally with your recommendation and reasoning
- Lead with your recommended option and explain why
Presenting the design:
- Once you believe you understand what you're building, present the design
- Scale each section to its complexity: a few sentences if straightforward, up to 200-300 words if nuanced
- Ask after each section whether it looks right so far
- Cover: architecture, components, data flow, error handling, testing
- Be ready to go back and clarify if something doesn't make sense
Design for isolation and clarity:
- Break the system into smaller units that each have one clear purpose, communicate through well-defined interfaces, and can be understood and tested independently
- For each unit, you should be able to answer: what does it do, how do you use it, and what does it depend on?
- Can someone understand what a unit does without reading its internals? Can you change the internals without breaking consumers? If not, the boundaries need work.
- Smaller, well-bounded units are also easier for you to work with - you reason better about code you can hold in context at once, and your edits are more reliable when files are focused. When a file grows large, that's often a signal that it's doing too much.
Working in existing codebases:
- Explore the current structure before proposing changes. Follow existing patterns.
- Where existing code has problems that affect the work (e.g., a file that's grown too large, unclear boundaries, tangled responsibilities), include targeted improvements as part of the design - the way a good developer improves code they're working in.
- Don't propose unrelated refactoring. Stay focused on what serves the current goal.
After the Design
Documentation:
- Write the validated design only when the task benefits from a durable specification.
- Follow the repository's documentation location and conventions.
- Do not commit unless the user requested a commit or repository instructions require it.
Spec Self-Review:
After writing the spec document, look at it with fresh eyes:
- Placeholder scan: Any "TBD", "TODO", incomplete sections, or vague requirements? Fix them.
- Internal consistency: Do any sections contradict each other? Does the architecture match the feature descriptions?
- Scope check: Is this focused enough for a single implementation plan, or does it need decomposition?
- Ambiguity check: Could any requirement be interpreted two different ways? If so, pick one and make it explicit.
Fix any issues inline. No need to re-review — just fix and move on.
User Review Gate:
After the spec review loop passes, ask the user to review the written spec before proceeding:
"Spec written and committed to <path>. Please review it and let me know if you want to make any changes before we start writing out the implementation plan."
Wait for the user's response. If they request changes, make them and re-run the spec review loop. Only proceed once the user approves.
Implementation:
- Create a detailed implementation plan using available planning capabilities.
- Use other relevant skills when their trigger conditions are satisfied.
Key Principles
- One question at a time - Don't overwhelm with multiple questions
- Multiple choice preferred - Easier to answer than open-ended when possible
- YAGNI ruthlessly - Remove unnecessary features from all designs
- Explore alternatives - Always propose 2-3 approaches before settling
- Incremental validation - Present design, get approval before moving on
- Be flexible - Go back and clarify when something doesn't make sense
1---2name: brainstorming3description: Explore requirements and compare design alternatives before substantial or ambiguous feature development. Use when the user asks to brainstorm a feature, define requirements, choose architecture, or plan a non-trivial behavioral change. Do not use for small, well-specified fixes or routine configuration changes.4---56# Brainstorming Ideas Into Designs78Help turn ideas into fully formed designs and specs through natural collaborative dialogue.910Start by understanding the current project context, then ask only questions whose11answers materially affect the design. Present a design and obtain approval before12implementing substantial or ambiguous changes. Skip this workflow for small,13well-specified fixes and routine configuration changes.1415## Checklist1617You MUST create a task for each of these items and complete them in order:18191. **Explore project context** — check files, docs, recent commits202. **Ask clarifying questions** — one at a time, understand purpose/constraints/success criteria213. **Propose 2-3 approaches** — with trade-offs and your recommendation224. **Present design** — in sections scaled to their complexity and obtain approval235. **Write a design document when useful** — use the repository's preferred location246. **Spec self-review** — check placeholders, contradictions, ambiguity, and scope257. **Transition to implementation** — create a plan appropriate to the task2627## Process Flow2829```dot30digraph brainstorming {31 "Explore project context" [shape=box];32 "Ask clarifying questions" [shape=box];33 "Propose 2-3 approaches" [shape=box];34 "Present design sections" [shape=box];35 "User approves design?" [shape=diamond];36 "Write design doc" [shape=box];37 "Spec self-review\n(fix inline)" [shape=box];38 "User reviews spec?" [shape=diamond];39 "Invoke writing-plans skill" [shape=doublecircle];4041 "Explore project context" -> "Ask clarifying questions";42 "Ask clarifying questions" -> "Propose 2-3 approaches";43 "Propose 2-3 approaches" -> "Present design sections";44 "Present design sections" -> "User approves design?";45 "User approves design?" -> "Present design sections" [label="no, revise"];46 "User approves design?" -> "Write design doc" [label="yes"];47 "Write design doc" -> "Spec self-review\n(fix inline)";48 "Spec self-review\n(fix inline)" -> "User reviews spec?";49 "User reviews spec?" -> "Write design doc" [label="changes requested"];50 "User reviews spec?" -> "Invoke writing-plans skill" [label="approved"];51}52```5354The terminal state is an approved design with a concrete implementation plan.5556## The Process5758**Understanding the idea:**5960- Check out the current project state first (files, docs, recent commits)61- Before asking detailed questions, assess scope: if the request describes multiple independent subsystems (e.g., "build a platform with chat, file storage, billing, and analytics"), flag this immediately. Don't spend questions refining details of a project that needs to be decomposed first.62- If the project is too large for a single spec, help the user decompose into sub-projects: what are the independent pieces, how do they relate, what order should they be built? Then brainstorm the first sub-project through the normal design flow. Each sub-project gets its own spec → plan → implementation cycle.63- For appropriately-scoped projects, ask questions one at a time to refine the idea64- Prefer multiple choice questions when possible, but open-ended is fine too65- Only one question per message - if a topic needs more exploration, break it into multiple questions66- Focus on understanding: purpose, constraints, success criteria6768**Exploring approaches:**6970- Propose 2-3 different approaches with trade-offs71- Present options conversationally with your recommendation and reasoning72- Lead with your recommended option and explain why7374**Presenting the design:**7576- Once you believe you understand what you're building, present the design77- Scale each section to its complexity: a few sentences if straightforward, up to 200-300 words if nuanced78- Ask after each section whether it looks right so far79- Cover: architecture, components, data flow, error handling, testing80- Be ready to go back and clarify if something doesn't make sense8182**Design for isolation and clarity:**8384- Break the system into smaller units that each have one clear purpose, communicate through well-defined interfaces, and can be understood and tested independently85- For each unit, you should be able to answer: what does it do, how do you use it, and what does it depend on?86- Can someone understand what a unit does without reading its internals? Can you change the internals without breaking consumers? If not, the boundaries need work.87- Smaller, well-bounded units are also easier for you to work with - you reason better about code you can hold in context at once, and your edits are more reliable when files are focused. When a file grows large, that's often a signal that it's doing too much.8889**Working in existing codebases:**9091- Explore the current structure before proposing changes. Follow existing patterns.92- Where existing code has problems that affect the work (e.g., a file that's grown too large, unclear boundaries, tangled responsibilities), include targeted improvements as part of the design - the way a good developer improves code they're working in.93- Don't propose unrelated refactoring. Stay focused on what serves the current goal.9495## After the Design9697**Documentation:**9899- Write the validated design only when the task benefits from a durable specification.100- Follow the repository's documentation location and conventions.101- Do not commit unless the user requested a commit or repository instructions require it.102103**Spec Self-Review:**104After writing the spec document, look at it with fresh eyes:1051061. **Placeholder scan:** Any "TBD", "TODO", incomplete sections, or vague requirements? Fix them.1072. **Internal consistency:** Do any sections contradict each other? Does the architecture match the feature descriptions?1083. **Scope check:** Is this focused enough for a single implementation plan, or does it need decomposition?1094. **Ambiguity check:** Could any requirement be interpreted two different ways? If so, pick one and make it explicit.110111Fix any issues inline. No need to re-review — just fix and move on.112113**User Review Gate:**114After the spec review loop passes, ask the user to review the written spec before proceeding:115116> "Spec written and committed to `<path>`. Please review it and let me know if you want to make any changes before we start writing out the implementation plan."117118Wait for the user's response. If they request changes, make them and re-run the spec review loop. Only proceed once the user approves.119120**Implementation:**121122- Create a detailed implementation plan using available planning capabilities.123- Use other relevant skills when their trigger conditions are satisfied.124125## Key Principles126127- **One question at a time** - Don't overwhelm with multiple questions128- **Multiple choice preferred** - Easier to answer than open-ended when possible129- **YAGNI ruthlessly** - Remove unnecessary features from all designs130- **Explore alternatives** - Always propose 2-3 approaches before settling131- **Incremental validation** - Present design, get approval before moving on132- **Be flexible** - Go back and clarify when something doesn't make sense