Decision Matrix Builder

# Decision Matrix Builder

Structure complex decisions with weighted scoring matrices, research-backed option analysis, and clear recommendations.

## When to Use

- "Help me decide between {option A} and {option B}"
- "Should I {choice}?"
- "I'm torn between 3 options — help me think through it"

## Requirements

- **Brave Search MCP** for researching options
- **python3** for weighted scoring calculations

## Workflow

### Step 1 — Frame the Decision

- **Decision** (what are you choosing between?)
- **Options** (2-5 concrete choices)
- **Criteria** (what matters to you — cost, time, quality, risk, etc.)
- **Context** (timeline, constraints, who is affected)

### Step 2 — Research Options

```
brave_web_search: "{OPTION_A} vs {OPTION_B} comparison {YEAR}"
brave_web_search: "{OPTION_A} pros cons review"
brave_web_search: "{OPTION_B} pros cons review"
```

### Step 3 — Weight Criteria

Ask user to rank criteria importance (or suggest defaults):

```bash
python3 << 'PYEOF'
criteria = {
    "{Criterion 1}": {WEIGHT_1},  # 1-10 importance
    "{Criterion 2}": {WEIGHT_2},
    "{Criterion 3}": {WEIGHT_3},
    "{Criterion 4}": {WEIGHT_4},
    "{Criterion 5}": {WEIGHT_5},
}

# Normalize weights to 100%
total = sum(criteria.values())
for k in criteria:
    criteria[k] = criteria[k] / total
    print(f"{k}: {criteria[k]*100:.0f}%")
PYEOF
```

### Step 4 — Score Options

```bash
python3 << 'PYEOF'
criteria_weights = {
    "{Criterion 1}": {NORMALIZED_W1},
    "{Criterion 2}": {NORMALIZED_W2},
    "{Criterion 3}": {NORMALIZED_W3},
    "{Criterion 4}": {NORMALIZED_W4},
    "{Criterion 5}": {NORMALIZED_W5},
}

# Score each option 1-10 per criterion
options = {
    "{Option A}": {{CRITERION_1}: {SCORE}, {CRITERION_2}: {SCORE}, ...},
    "{Option B}": {{CRITERION_1}: {SCORE}, {CRITERION_2}: {SCORE}, ...},
    "{Option C}": {{CRITERION_1}: {SCORE}, {CRITERION_2}: {SCORE}, ...},
}

print(f"{'Option':<20} {'Weighted Score':>15} {'Rank':>6}")
print("=" * 45)

results = []
for name, scores in options.items():
    weighted = sum(scores[c] * criteria_weights[c] for c in criteria_weights)
    results.append((name, weighted))

results.sort(key=lambda x: -x[1])
for rank, (name, score) in enumerate(results, 1):
    marker = " ← Winner" if rank == 1 else ""
    print(f"{name:<20} {score:>15.2f} {rank:>5}{marker}")
PYEOF
```

### Step 5 — Present Decision Matrix

```markdown
# Decision: {DECISION}

## Scoring Matrix
| Criterion (Weight) | {Option A} | {Option B} | {Option C} |
|--------------------|------------|------------|------------|
| {C1} ({W1}%) | {score}/10 | {score}/10 | {score}/10 |
| {C2} ({W2}%) | {score}/10 | {score}/10 | {score}/10 |
| {C3} ({W3}%) | {score}/10 | {score}/10 | {score}/10 |
| **Weighted Total** | **{X.XX}** | **{X.XX}** | **{X.XX}** |
| **Rank** | {#} | {#} | {#} |

## 🏆 Recommendation: {WINNER}
{2-3 sentences explaining why this option wins and key trade-offs}

## Trade-offs to Consider
- {Winner} scores lower on {criterion} — is that acceptable?
- {Runner-up} is close — wins if you value {criterion} more

## Gut Check
Does this result match your intuition? If not, which criterion needs re-weighting?
```

## Important Rules

- Present the framework, not just the answer — help the user think
- Scores must be justified with research or facts
- Include the "gut check" — math doesn't capture everything
- If options score within 5% of each other, say it's a toss-up
- Don't make the decision for the user — present the analysis

## Example Prompts

- "Help me decide between staying at my job vs. taking a new offer"
- "Should I buy or rent? Help me think through it"
- "Compare AWS vs GCP vs Azure for our startup"