You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, designing systems, or making architectural decisions. Enters plan mode, reads all available docs, explores the codebase deeply, then interviews the user relentlessly with ultrathink-level reasoning on every decision until a shared understanding is reached. Produces a validated design spec before any implementation begins. Triggers on feature requests, design discussions, refactors, new projects, component creation, system changes, and any task requiring design decisions.
Instrucciones de origen · Vista previa de solo lectura
name
super-brainstorm
version
0.1.0
description
You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, designing systems, or making architectural decisions. Enters plan mode, reads all available docs, explores the codebase deeply, then interviews the user relentlessly with ultrathink-level reasoning on every decision until a shared understanding is reached. Produces a validated design spec before any implementation begins. Triggers on feature requests, design discussions, refactors, new projects, component creation, system changes, and any task requiring design decisions.
[{"url":"https://github.com/obra/superpowers/tree/main/skills/brainstorming","accessed":"2026-03-18T00:00:00.000Z","description":"Original brainstorming skill from obra/superpowers - interview-driven design process with spec review loop"},{"url":"https://github.com/AbsolutelySkilled/AbsolutelySkilled","accessed":"2026-03-18T00:00:00.000Z","description":"Enhanced with ultrathink reasoning, codebase-first intelligence, strict dependency-resolved interviews, and flexible exit"}]
license
MIT
maintainers
[{"github":"maddhruv"}]
When this skill is activated, always start your first response with the brain emoji.
Super Brainstorm
Activation Banner
At the very start of every super-brainstorm invocation, before any other output, display this ASCII art banner:
Follow the banner immediately with: Entering plan mode - ultrathink enabled
A relentless, ultrathink-powered design interview that turns vague ideas into
bulletproof specs. This is not a casual brainstorm - it is a structured
interrogation of every assumption, every dependency, and every design branch
until the AI and user reach a shared understanding that a staff engineer would
approve.
When to use this skill
Trigger this skill when the user:
Requests a new feature, component, or functionality that needs design before implementation
Asks to build something and the scope or approach is not yet clear
Wants to brainstorm or explore options for a design decision
Is starting a greenfield project and needs to define architecture
Needs to refactor or redesign an existing system
Says "let's think through this" or "help me plan this"
Wants to decompose a complex request into a validated spec before coding
Do NOT trigger this skill for:
Quick bug fixes or typo corrections where the solution is obvious
Pure code review tasks with no design decisions
Tasks where the user explicitly says "just do it" and the scope is trivially clear
Hard Gates
1. ALWAYS enter plan mode at the start of this skill. Do not proceed in normal mode.
2. ALWAYS use extended thinking (ultrathink / think hard) before every question,
every decision, every approach proposal, and every design section. No exceptions.
3. Do NOT invoke any implementation skill, write any code, scaffold any project,
or take any implementation action until the spec is written and the user has
approved it. This applies to EVERY project regardless of perceived simplicity.
Anti-Pattern: "This Is Too Simple To Need A Design"
Every project goes through this process. A todo list, a single-function utility,
a config change - all of them. "Simple" projects are where unexamined assumptions
cause the most wasted work. The design can be short (a few sentences for truly
simple projects), but you MUST present it and get approval.
Existing patterns, conventions, and architectural decisions
Tech stack and dependencies
Testing patterns and CI setup
Overlapping or related design docs
Code style and organizational conventions
Output to the user: A brief summary of what you found, highlighting anything
relevant to the task at hand. Do NOT dump a file listing - synthesize what matters.
Phase 2: Codebase-First Intelligence
Before asking ANY question, check if the codebase already answers it.
This is the core differentiator. The AI must:
Identify what it needs to know to make the next design decision
Search the codebase for the answer (grep, glob, read relevant files)
Only ask the user if the code genuinely cannot answer the question
Examples:
"What database are you using?" - DON'T ASK. Check package.json, config files, existing code.
"How do you handle authentication?" - DON'T ASK. Search for auth middleware, JWT usage, session handling.
"What's the visual style you want?" - ASK. Code can't answer aesthetic preferences.
"Should this be real-time or batch?" - ASK. This is a product decision the codebase can't resolve.
When you DO find the answer in the codebase, tell the user what you found:
"I see you're using Prisma with PostgreSQL (from prisma/schema.prisma). I'll design around that."
This builds confidence and saves the user from answering questions they already answered in code.
Phase 3: Scope Assessment
Before diving into 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
Help decompose into sub-projects: what are the independent pieces, how do they
relate, what order should they be built?
Each sub-project gets its own brainstorm -> spec -> plan -> implementation cycle
Brainstorm the first sub-project through the normal design flow
If the request is appropriately scoped, proceed to the interview.
Phase 4: Relentless Interview
This is the heart of the skill. Walk down every branch of the design tree,
resolving dependencies between decisions one by one.
Rules:
Use the AskUserQuestion tool for every question - this is a built-in Claude
Code tool that pauses execution and waits for the user's response. Use it for
every interview question, every section approval, and every decision point. Never
just print a question in your output - always use the tool so the conversation
properly blocks until the user responds.
One question at a time - never overwhelm with multiple questions
Ultrathink before every question - reason deeply about what you need to know
next, what depends on what, and whether the codebase can answer it
Strictly linear - if decision B depends on decision A, never ask about B
until A is locked
Prefer multiple choice when possible - easier to answer. Include your
recommendation marked as (Recommended) with a clear rationale
Never fake options - only propose multiple approaches when there's a genuine
fork in the road. If the codebase and interview clearly point to one right
answer, present it with reasoning for why alternatives were dismissed
Codebase check before every question - search the code first, only ask what
code can't tell you
Keep going until every decision node is resolved - don't shortcut, don't
assume, don't hand-wave. If a branch of the design tree hasn't been explored,
explore it.
What to interview about:
Purpose and success criteria
User/consumer personas
Data model and relationships
Component boundaries and interfaces
State management and data flow
Error handling and edge cases
Performance requirements and constraints
Security considerations
Testing strategy
Migration path (if modifying existing systems)
Backwards compatibility concerns
Design tree traversal:
Think of the design as a tree of decisions. Each decision may open new branches.
Walk the tree depth-first, resolving each branch fully before moving to siblings.
Feature X
- Who is this for? (resolve)
- What's the core interaction? (resolve)
- How does data flow? (resolve)
- What are the edge cases? (resolve)
- What are the error states? (resolve)
- What's the secondary interaction? (resolve)
- How does this integrate with existing system? (resolve)
Phase 5: Approach Proposals
Only propose multiple approaches when there is a genuine design fork.
When the answer is obvious:
Present the single approach with reasoning. Briefly mention why you dismissed
alternatives:
"Given your existing Express + Prisma stack and the read-heavy access pattern,
a new Prisma model with a cached read path is the clear approach. A separate
microservice would add complexity without benefit at this scale, and a raw SQL
approach would lose Prisma's type safety."
When there's a genuine fork:
Present each option with:
What it is (1-2 sentences)
Pros and cons
When you'd pick it
Mark one as (Recommended) with clear rationale
Phase 6: Design Presentation
Once the design tree is fully resolved, present the design section by section.
Rules:
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 approach
Be ready to go back and revise if something doesn't fit
Reference existing codebase patterns you're building on
Design for isolation and clarity:
Break the system into smaller units with one clear purpose each
Well-defined interfaces between units
Each unit should be understandable and testable independently
For each unit: what does it do, how do you use it, what does it depend on?
Working in existing codebases:
Follow existing patterns. Don't fight the codebase.
Where existing code has problems that affect the work (e.g., a file that's grown
too large, unclear boundaries), 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.
Phase 7: Write Spec
After user approves the full design:
Write to docs/plans/YYYY-MM-DD-<topic>-design.md
Clear, concise prose. No fluff.
Sections should mirror what was discussed and approved
After writing the spec, dispatch a reviewer subagent:
Agent tool (general-purpose):
description: "Review spec document"
prompt: |
You are a spec document reviewer. Verify this spec is complete and ready
for implementation planning.
Spec to review: [SPEC_FILE_PATH]
| Category | What to Look For |
|------------- |---------------------------------------------------------------|
| Completeness | TODOs, placeholders, "TBD", incomplete sections |
| Consistency | Internal contradictions, conflicting requirements |
| Clarity | Requirements ambiguous enough to cause building the wrong thing|
| Scope | Focused enough for a single plan |
| YAGNI | Unrequested features, over-engineering |
Only flag issues that would cause real problems during implementation.
Approve unless there are serious gaps.
Output format:
## Spec Review
**Status:** Approved | Issues Found
**Issues (if any):**
- [Section X]: [specific issue] - [why it matters]
**Recommendations (advisory, do not block approval):**
- [suggestions]
If issues found: fix them, re-dispatch, repeat
Max 3 iterations. If still failing, surface to the user for guidance.
Phase 9: User Reviews Spec
After the review loop passes:
"Spec written to <path>. Please review it and let me know if you want to make
any changes before we proceed."
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.
Phase 10: Flexible Exit
Once the spec is approved, present the user with options:
"Spec is approved. What would you like to do next?"
A) Writing plans - create a detailed implementation plan (invoke writing-plans skill)
B) Superhuman - full AI-native SDLC with task decomposition and parallel execution (invoke super-human skill)
C) Direct implementation - start building right away
D) Something else - your call
Let the user decide the next step. Do not auto-invoke any skill.
Key Principles
Ultrathink everything - deep reasoning before every decision, no lazy shortcuts
Codebase before questions - respect the user's time, only ask genuine unknowns
One question at a time via AskUserQuestion tool - never overwhelm, always use the built-in tool to ask
Strictly linear - resolve dependencies before moving forward
Honest options - real forks get multiple approaches, obvious answers get presented directly
Always mark (Recommended) - every set of options includes a clear recommendation with rationale
YAGNI ruthlessly - remove unnecessary features from all designs
Incremental validation - present design section by section, get approval before moving on
Plan mode always - this skill operates entirely in plan mode
Gotchas
AskUserQuestion tool not available in all environments - The AskUserQuestion tool is a Claude Code-specific built-in. In other environments (Gemini CLI, OpenAI Codex), it may not exist. Fall back to printing the question as a clearly demarcated output block and waiting for user response, but track that you are waiting for an answer before proceeding.
Deep context scan can consume the entire context window - Reading every file in docs/ and every recent commit in a large codebase can exhaust the context window before the first question is asked. Be selective: read README, CLAUDE.md, and recent commits first; only go deeper on files directly relevant to the task.
Spec saved to docs/plans/ in the wrong repo - If the skill is invoked in a monorepo or a workspace with multiple docs/ directories, saving the spec to the wrong subdirectory means it will never be found during future DISCOVER phases. Confirm the target docs/plans/ path with the user before writing.
Reviewer subagent approves incomplete specs - The reviewer subagent is prompted to "approve unless there are serious gaps," which means minor incompleteness often passes. Do not treat reviewer approval as a substitute for user approval. The user gate in Phase 9 is mandatory regardless of the reviewer's verdict.
Flexible exit auto-invokes the next skill - Presenting the exit options and then immediately invoking a skill without waiting for user input defeats the purpose of a flexible exit. Always use AskUserQuestion (or equivalent) to receive the user's choice before taking any post-spec action.
Anti-Patterns and Common Mistakes
Anti-Pattern
Better Approach
Asking questions the codebase can answer
Search code first - check configs, existing patterns, test files before every question
Batching multiple questions in one message
One question at a time, always. Break complex topics into sequential questions
Printing questions as plain text output
Always use the AskUserQuestion tool to ask - it blocks until the user responds
Skipping docs/ and README before starting
Always read all available documentation before the first question
Proposing fake alternatives when the answer is obvious
Present the single right answer with rationale; only show options at genuine forks
Accepting vague answers without follow-up
Dig deeper - "what do you mean by that?" is always valid
Asking about implementation before purpose
Always resolve "why" and "what" before "how"
Not exploring error/edge case branches
Every design tree has an error handling branch - walk it
Jumping to code before spec approval
Hard gate: no code, no scaffolding, no implementation until spec is approved
Presenting options without a (Recommended) marker
Every option set must have a clear recommendation with rationale
Using normal thinking when ultrathink is required
Ultrathink on every decision, every question, every proposal - no exceptions
Decomposing too late
Flag multi-system scope immediately, don't spend 10 questions refining details of an unscoped project
Auto-invoking the next skill without asking
Flexible exit - always let the user choose what happens after spec approval
References
For detailed guidance on specific aspects, load these reference files:
references/interview-playbook.md - Design tree traversal, question banks by project type, codebase-first intelligence patterns, example interview sessions
references/spec-writing.md - Spec document template, section scaling rules, writing style guide, decision log format, spec review checklist, example spec
references/approach-analysis.md - When to propose multiple approaches, approach proposal format, trade-off dimensions, project decomposition guide, common decision trees
Only load a references file if the current phase requires it - they are long and will consume context.
Companion check
On first activation of this skill in a conversation: check which companion skills are installed by running ls ~/.claude/skills/ ~/.agent/skills/ ~/.agents/skills/ .claude/skills/ .agent/skills/ .agents/skills/ 2>/dev/null. Compare the results against the recommended_skills field in this file's frontmatter. For any that are missing, mention them once and offer to install:
