/find

Find Brand Names

Parse the user input to extract:

• Description: What the company/product does, its mission, target audience
• Ideas (optional): User-provided name candidates after "ideas:" to include in evaluation

Workflow Overview

This is a multi-stage funnel that efficiently narrows candidates:

Brainstorm (15-20) → Quick Filter (~8-10) → [User checkpoint]
    → Conflict Check (~4-6) → Deep Evaluation → Final Ranking (Top 3)

---

Stage 1: Brainstorm

Generate 15-20 brand name candidates. Mix of styles:

• Abstract/evocative: Suggest without describing (Stripe, Notion)
• Portmanteaus: Blended words (Instagram, Pinterest)
• Latin/Greek roots: Scientific feel (Anthropic, Veritas)
• Simple real words: Unexpected contexts (Apple, Slack)
• Invented words: Completely new (Kodak, Xerox)

If user provided "ideas:", include those in the candidate list.

Output:

BRAINSTORM: Generated 18 candidates
[list each with one-line rationale]

---

Stage 2: Quick Filter (Parallel, Cheap)

Run these checks for ALL candidates simultaneously:

2a. Domain Availability (Bash)

# Check .com
whois {name}.com 2>/dev/null | head -20

# Check .ai (use specific server)
whois -h whois.nic.ai {name}.ai 2>/dev/null | head -10

• "No match" / "NOT FOUND" / "Domain not found" = available
• Registration dates present = taken

2b. Phonetic Analysis (Optional)

python -c "from big_phoney import BigPhoney" 2>/dev/null && python -m namecast.phonetic "{name}" || echo "SKIP"

Flag issues:

• sounds_like containing confusing words ("sock", "suck", etc.)
• difficulty = "hard"
• Syllables > 4

2c. Basic Heuristics

• Length: 4-10 characters ideal, >12 = flag
• No numbers or hyphens
• Can spell from hearing it once

Elimination criteria (any = out):

• Both .com AND .ai taken
• Sounds like offensive/confusing word

• 12 characters

---

Stage 3: Checkpoint - Show Results + Offer Choice

IMPORTANT: Stop here and show user what passed/failed before continuing.

## Quick Filter Results

### Passed: {N} names
| Name | .com | .ai | Syllables | Notes |
|------|------|-----|-----------|-------|
| Cosilico | ✓ | owned | 4 | Clear pronunciation |
| ... | ... | ... | ... | ... |

### Eliminated: {M} names
| Name | Reason |
|------|--------|
| ModelSoc | Phonetic: sounds like "model sock" |
| TaxCloud | .com and .ai both taken (active sites) |
| Economicus | 5 syllables, hard to remember |

---

**What would you like to do?**
1. **Continue** → Deep evaluation of the {N} passed names
2. **Refine** → Adjust your description and I'll brainstorm again
3. **Add names** → Tell me specific names to add to the evaluation

Wait for user response before proceeding.

---

Stage 4: Conflict Check

For survivors, check for existing companies and international issues:

4a. Similar Companies (WebSearch)

Search: "{name}" company OR startup OR brand

Classify each result:

• 🔴 Active competitor (same market, live site) → eliminate
• 🟡 Tangential (different market/geography) → flag only
• ⚪ Defunct (no live site, old mentions) → ignore

4b. International Meanings

Check major languages for problematic meanings:

• Spanish, French, German, Mandarin, Japanese, Portuguese, Arabic

Elimination criteria:

• Active same-market competitor with similar name
• Offensive/negative meaning in major language

---

Stage 5: Deep Evaluation

For finalists (target 4-6 names):

5a. Infer Target Personas

Based on description, identify 5 relevant personas:

Example for B2B API company:
1. Enterprise CTO (50s) - risk-averse, cares about stability
2. Startup founder (30s) - moves fast, values developer experience
3. Developer advocate (28) - evaluates technical credibility
4. VC partner (40s) - pattern-matches against portfolio
5. Government procurement (45) - needs "safe" vendors

5b. Persona Perception Survey

For each persona × name, roleplay and rate:

• Memorability (1-10)
• Professionalism (1-10)
• Trust factor (1-10)
• One-sentence gut reaction
• Would use/recommend? (yes/maybe/no)

5c. Brand Scope Analysis

Rate each name (1-10 scale):

• Narrowness: Does name box in the company? (10 = unlimited like "Amazon")
• Expansion potential: Can grow into adjacent areas?
• Mission alignment: Captures full vision, not just current product?

5d. Tagline Generation

For each name, generate 3 taglines that:

• Complement/explain the name
• Capture the full mission
• Are memorable and quotable

---

Stage 6: Final Ranking

Scoring (0-100 points)

Category	Points	Criteria
Domain availability	20	.com avail=10, .ai avail=10
Brand scope	20	Scope rating × 2
Persona perception	25	Avg rating across personas, scaled
Pronunciation	15	Easy=15, Medium=10, Hard=5 (skip if unavailable)
Linguistic quality	10	Spelling ease, memorability
Uniqueness	10	No conflicts=10, flagged=5, competitor=0

Present Results

## 🏆 Top 3 Recommendations

### 1. [NAME] — Score: XX/100

**Quick facts:**
- Domains: .com ✓ / .ai ✓
- Pronunciation: X syllables, [easy/medium/hard]
- Scope: X/10 — "[assessment]"

**Persona reactions:**
| Persona | Rating | Reaction |
|---------|--------|----------|
| CTO (50s) | 8/10 | "Sounds enterprise-ready" |
| ... | ... | ... |

**Suggested taglines:**
1. "[tagline]"
2. "[tagline]"
3. "[tagline]"

---

### 2. [NAME] — Score: XX/100
[same format]

### 3. [NAME] — Score: XX/100
[same format]

---

## Full Comparison

| Name | Domain | Scope | Perception | Pronunciation | Uniqueness | **Total** |
|------|--------|-------|------------|---------------|------------|-----------|
| ... | X/20 | X/20 | X/25 | X/15 | X/10 | **XX/100** |

---

## Recommendation

**[Winner]** is the strongest choice because:
- [Key strength 1]
- [Key strength 2]
- [Key strength 3]

**Consider [Runner-up] if:** [specific condition where it's better]

**Avoid [Eliminated name] because:** [key issue]

---

*Disclaimer: This is general information, not legal advice. Consult a trademark attorney before finalizing your brand name.*