To run this task you must have the following required information:

> Existing brainstorming session with accumulated ideas

If you don't have all of this information, exit here and respond asking for any extra information you require, and instructions to run this task again with ALL required information.

---

You MUST use a todo list to complete these steps in order. Never move on to one step if you haven't completed the previous step. If you have multiple read steps in a row, read them all at once (in parallel).

Add all steps to your todo list now and begin executing.

## Steps

1. SELECT SESSION

Read `documents/brainstorm-sessions/*.md`.

If multiple "exploring" sessions: ask which to evaluate.
If one "exploring" session: use it, confirm topic and idea count.
If no "exploring" but converged exist: offer to review one or start fresh.
If no sessions: "No brainstorming sessions found. Want to start one?"


2. HANDLE CONVERGED SESSIONS

If user selects an already-converged session, show their prior selections and offer:
- Add more ideas (reopen for brainstorming)
- Re-evaluate with different criteria (proceed to grouping below)
- Just review (show selections and stop)

Route accordingly before continuing.


3. [Read Brainstorming Evaluation]: Read the documentation in: `skills/sauna/[skill_id]/references/thinking.brainstorm.evaluation.md` (Load evaluation criteria and process)

4. GROUP IDEAS BY THEME

Read through all accumulated ideas and cluster them into 3-6 natural groupings based on:
- Similar mechanisms (ideas that work the same way)
- Same problem area (ideas that address the same root cause)
- Resource requirements (quick wins vs. big bets)
- Target audience (ideas for different user segments)

Give each cluster a descriptive theme name.
Note any ideas that don't fit cleanly — these might be the most novel.


5. PRESENT THEME OVERVIEW

Show the user the groupings:

"I've clustered your [N] ideas into [M] themes:

**[Theme 1]** ([X] ideas): [Brief description of what these ideas share]
**[Theme 2]** ([Y] ideas): [Brief description]
...

A few ideas don't fit neatly — these might be your most unique ones:
- [Outlier idea 1]
- [Outlier idea 2]"


6. SURFACE TOP CANDIDATES

For each theme, highlight 1-2 standout ideas and explain why they stand out:

"**From [Theme 1]:**
- **[Idea title]**: [Why this stands out — most impactful, most feasible, most novel, etc.]

**From [Theme 2]:**
- **[Idea title]**: [Why this stands out]"

Include any compelling outliers too.


7. COMPARE FINALISTS

Create a simple comparison of the top 4-6 ideas:

| Idea | Impact | Feasibility | Notes |
|------|--------|-------------|-------|
| [Idea 1] | High | Medium | [Key tradeoff or consideration] |
| [Idea 2] | Medium | High | [Key tradeoff or consideration] |
...

Ask if the user has different criteria that matter more:
"I'm evaluating on impact and feasibility. Are there other criteria that matter more for your situation? (e.g., speed, cost, alignment with strategy)"


8. INVITE USER CHOICE

Based on the comparison:
"**[Idea X]** and **[Idea Y]** seem strongest because [brief rationale].

Which resonates most with you? Or is there one from the list that I'm undervaluing?"

Let the user select 1-3 ideas to pursue.


9. ELABORATE WINNERS

For each idea the user selects, provide:

**What it is:** Clear description in 2-3 sentences

**Why it works:** The core insight or mechanism that makes this promising

**First steps:** 2-3 concrete actions to start exploring this idea

**Open questions:** What would need to be true for this to succeed? What's the biggest unknown?

**Risks:** What could go wrong? What's the main objection?


10. CLOSE SESSION

Update the session file:
1. Change status from "exploring" to "converged"
2. Add a "## Final Selection" section with the chosen ideas and elaboration
3. Keep all the original ideas intact for future reference

Summarize for the user:
"We explored [topic] from [N] angles and generated [M] ideas. You've selected:

- **[Idea 1]**: [One-liner]
- **[Idea 2]**: [One-liner]

This session is saved — you can revisit it anytime by asking to 'continue my brainstorm about [topic]' or 'show my brainstorm about [topic].'"


11. IF USER CAN'T DECIDE

If the user struggles to choose:

**Too many good options:** "These are all strong. What if we picked one to start with? We can always return to the others."

**Nothing feels right:** "It sounds like we haven't hit the right angle yet. Want to do another brainstorming round with a different technique?"

**Analysis paralysis:** "What's the smallest version of any of these we could try this week?"