Brainstorming Ideas Into Designs
Overview
Help turn ideas into fully formed designs and specs through natural collaborative dialogue.
Start by understanding the current project context, 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.
The Process
Understanding the idea:
- Check out the current project state first (files, docs, recent commits)
- 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
- Break it into sections of 200-300 words
- Ask after each section whether it looks right so far
- Cover: architecture, components, data flow, error handling, state behavior, edge cases, testing
- Be ready to go back and clarify if something doesn't make sense
Validating for testability (BDD readiness check):
After the full design is presented and validated, run every design point through this checklist before finalizing:
| Dimension |
Question to verify |
| Input/Output format |
Are the exact formats of inputs and outputs specified? (data types, structure, encoding) |
| Error & exception scenarios |
Is every failure mode explicitly described with its expected behavior? (not just the happy path) |
| Boundary & priority rules |
When ambiguity or conflict can arise, are the resolution rules defined? (precedence, fallback, default values) |
| State behavior |
Is it clear what state persists, what is isolated, and what resets? (sessions, variables, side effects) |
| Verifiable granularity |
Can each behavior be independently tested with concrete steps and a single expected outcome? |
| Ambiguity check |
Are there any implicit assumptions that different readers could interpret differently? |
How to use the checklist:
- For each design section, evaluate all 6 dimensions
- Any dimension that fails → go back to the user with a targeted question to fill the gap
- Do NOT silently assume defaults — if the PRD will be consumed downstream (e.g., converted to BDD test cases), ambiguity is a defect
- Only proceed to documentation once all design points pass all 6 dimensions
- Mark dimensions as "N/A" only when genuinely not applicable (e.g., stateless operations have no state behavior)
After the Design
Documentation:
- Write the validated design to
docs/plans/YYYY-MM-DD-<topic>-design.md
- Use elements-of-style:writing-clearly-and-concisely skill if available
- Commit the design document to git
Implementation (if continuing):
- Ask: "Ready to set up for implementation?"
- Use superpowers:using-git-worktrees to create isolated workspace
- Use superpowers:writing-plans to create detailed implementation plan
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 in sections, validate each
- Be flexible - Go back and clarify when something doesn't make sense
1---2name: brainstorming-83description: 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.4---5
6# Brainstorming Ideas Into Designs
7
8## Overview
9
10Help turn ideas into fully formed designs and specs through natural collaborative dialogue.
11
12Start by understanding the current project context, 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.
13
14## The Process
15
16**Understanding the idea:**
17- Check out the current project state first (files, docs, recent commits)
18- Ask questions one at a time to refine the idea
19- Prefer multiple choice questions when possible, but open-ended is fine too
20- Only one question per message - if a topic needs more exploration, break it into multiple questions
21- Focus on understanding: purpose, constraints, success criteria
22
23**Exploring approaches:**
24- Propose 2-3 different approaches with trade-offs
25- Present options conversationally with your recommendation and reasoning
26- Lead with your recommended option and explain why
27
28**Presenting the design:**
29- Once you believe you understand what you're building, present the design
30- Break it into sections of 200-300 words
31- Ask after each section whether it looks right so far
32- Cover: architecture, components, data flow, error handling, state behavior, edge cases, testing
33- Be ready to go back and clarify if something doesn't make sense
34
35**Validating for testability (BDD readiness check):**
36
37After the full design is presented and validated, run every design point through this checklist before finalizing:
38
39| Dimension | Question to verify |
40|---|---|
41| Input/Output format | Are the exact formats of inputs and outputs specified? (data types, structure, encoding) |
42| Error & exception scenarios | Is every failure mode explicitly described with its expected behavior? (not just the happy path) |
43| Boundary & priority rules | When ambiguity or conflict can arise, are the resolution rules defined? (precedence, fallback, default values) |
44| State behavior | Is it clear what state persists, what is isolated, and what resets? (sessions, variables, side effects) |
45| Verifiable granularity | Can each behavior be independently tested with concrete steps and a single expected outcome? |
46| Ambiguity check | Are there any implicit assumptions that different readers could interpret differently? |
47
48**How to use the checklist:**
49- For each design section, evaluate all 6 dimensions
50- Any dimension that fails → go back to the user with a targeted question to fill the gap
51- Do NOT silently assume defaults — if the PRD will be consumed downstream (e.g., converted to BDD test cases), ambiguity is a defect
52- Only proceed to documentation once all design points pass all 6 dimensions
53- Mark dimensions as "N/A" only when genuinely not applicable (e.g., stateless operations have no state behavior)
54
55## After the Design
56
57**Documentation:**
58- Write the validated design to `docs/plans/YYYY-MM-DD-<topic>-design.md`
59- Use elements-of-style:writing-clearly-and-concisely skill if available
60- Commit the design document to git
61
62**Implementation (if continuing):**
63- Ask: "Ready to set up for implementation?"
64- Use superpowers:using-git-worktrees to create isolated workspace
65- Use superpowers:writing-plans to create detailed implementation plan
66
67## Key Principles
68
69- **One question at a time** - Don't overwhelm with multiple questions
70- **Multiple choice preferred** - Easier to answer than open-ended when possible
71- **YAGNI ruthlessly** - Remove unnecessary features from all designs
72- **Explore alternatives** - Always propose 2-3 approaches before settling
73- **Incremental validation** - Present design in sections, validate each
74- **Be flexible** - Go back and clarify when something doesn't make sense
majiayu000