skills/brainstorm/SKILL.md
Use when you need structured problem exploration before writing requirements.
npx skillsauth add seokan-jeong/team-shinchan team-shinchan:brainstormInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
If args is empty or only whitespace: Ask user: "What problem or feature would you like to brainstorm? Describe the goal, not the solution." STOP and wait for user response
If args length > 2000 characters: Truncate to 2000 characters Warn user: "Request was truncated to 2000 characters"
Check if .shinchan-docs/*/brainstorm-output.md exists for the current session.
If found: read it and include a note in the Hiroshi prompt: "A prior brainstorm exists. Extend or revise it rather than starting fresh."
Task( subagent_type="team-shinchan:hiroshi", model="opus", prompt=`/team-shinchan:brainstorm has been invoked.
Problem/Goal to explore: ${args}
You are doing structured problem exploration BEFORE requirements. The user may have arrived with a solution in mind ("add X feature"). Your job is to help them see the problem clearly before committing to a solution direction.
State the underlying problem in 2-3 sentences. Strip away solution framing. Example: "The user wants faster search" → "Users cannot find relevant items within their session window, causing abandonment."
Present 2-4 alternative approaches to address the problem. For each:
Option A: [Title]
Option B: [Title] [same structure]
[Option C, D if relevant]
State which option you recommend and why. Include:
User request: ${args} `)
After Hiroshi completes, if an active WORKFLOW_STATE.yaml exists:
Save Hiroshi's output to .shinchan-docs/{DOC_ID}/brainstorm-output.md
This allows /team-shinchan:requirements to reference it in Step 1.
Present the brainstorm output to the user. Then ask: "Ready to proceed with /team-shinchan:requirements? (This will start the full requirements interview based on the recommended path.)"
If user confirms: suggest running /team-shinchan:requirements with the recommended approach as context.
testing
Default-on interview option-quality panel — N diverse generators produce structure-free options, a SelfCheckGPT majority-vote consensus filters hallucinations, a SteerConf cautious-confidence judge scores survivors, and a deterministic top-K is returned. Workflow tier; the single fierce-* skill that is ON by default.
development
Deterministic adversarial code review for high-stakes scope — independent per-dimension review, a non-skippable per-finding refutation, completeness + interaction critics, and a deterministic 3-lens rubric judge panel. Opt-in main-loop Workflow tier.
data-ai
Deterministic loop-until-done for high-stakes long-running tasks — a worker/verifier loop the script bounds by iteration cap, token budget, and stagnation, closed by an Action-Kamen gate. Opt-in main-loop Workflow tier.
testing
Deterministic adversarial debate for high-stakes or irreversible decisions — mandatory refutation plus a scored judge panel. Opt-in main-loop Workflow tier.