# /vision — Product Vision Definition

## Usage
```
/vision                # Generate or refresh product vision
/vision refresh        # Force regenerate from latest positioning + business context
/vision score          # Score all approved ideas against vision alignment
```

## Input
- `$ARGUMENTS` — optional: "refresh" or "score"

## Process

### Step 1: Load Context
1. Read `data/product-vision.json` — existing vision (if any)
2. Read `data/positioning.json` — current positioning data
3. Read `data/business-context.json` — product info, funnels, monetization
4. Read `data/competitors.json` — competitor profiles
5. Determine mode:
   - No vision exists → full generation
   - `"refresh"` → regenerate treating existing vision as draft
   - `"score"` → score approved ideas against current vision

### Step 2: Generate Product Vision (skip if mode is "score")

Use two proven frameworks:

**Obviously Awesome (April Dunford) — 5-step positioning:**
1. Competitive alternatives — What would customers do if you didn't exist?
2. Unique attributes — What capabilities do you have that alternatives lack?
3. Value — What value do those attributes enable for customers?
4. Best-fit customers — Who cares most about that value?
5. Market category — What context makes your value obvious?

**StoryBrand (Donald Miller) — 7-part narrative:**
1. Character — The customer (not you) — who are they, what do they want?
2. Problem — External (tangible), Internal (emotional), Philosophical (why it matters)
3. Guide — You: empathy ("we understand") + authority ("we've done this")
4. Plan — 3 simple steps to succeed with your product
5. Call to Action — Direct CTA + transitional CTA
6. Failure — What happens if they don't act (stakes)
7. Success — The transformation (before → after)

Synthesize into:
- **One-liner**: "For [customers] who [problem], [product] is the [category] that [value]. Unlike [alternatives], we [unique_attributes]."
- **Focus metric**: The ONE metric that proves this vision is working
- **Anti-goals**: Things we explicitly will NOT build (prevents scope creep)

### Step 3: Save Vision
Write to `data/product-vision.json` with full schema.

### Step 4: Derive Hypotheses
From the vision, generate 5-8 testable business hypotheses:
- Each linked to a funnel stage (acquisition → engagement → conversion → monetization)
- Each falsifiable: "If we [action], then [outcome] because [reason]"
- Save to `data/hypotheses.json` (append, don't overwrite existing)

### Step 5: Score Ideas (if mode is "score")
For each approved idea in `data/ideas.json`:
- Calculate alignment score (0-100) against the vision
- Show results sorted by alignment
- Flag ideas that hit anti-goals

## Output
- Updated `data/product-vision.json`
- New hypotheses in `data/hypotheses.json` (if generated)
- Summary table showing vision canvases + one-liner + focus metric