naming

# Placek Naming Skill

An agent skill for naming companies, products, features, projects, and platforms using David Placek / Lexicon Branding–style methodology.

This is not a loose brainstorming prompt. It is a structured naming process built around product understanding, category contrast, linguistic filters, and the productive tension that makes names memorable.

## What it does

The skill guides an agent through:

1. **Product / category intake** — learn what is actually being named before asking category-relative questions.
2. **Comfort Trap Interview** — surface the user’s bias toward safe, descriptive names.
3. **Landscape audit** — map competitor/category naming conventions so the work can avoid them.
4. **Deep product understanding** — clarify users, buyers, advantages, and strategic naming goals.
5. **Ultimate benefit discovery** — climb from feature to feeling.
6. **Treasure hunt generation** — explore roots, mythology, adjacent domains, sound symbolism, blends, and distant associations.
7. **Linguistic filtering** — evaluate fluency, memorability, sound, and distinctiveness.
8. **Believability testing** — put names in real contexts instead of spreadsheets.
9. **Final evaluation** — score names for originality, surprise, searchability, cross-market risk, and compounding value.

## Files

```text
SKILL.md
references/
  roots-and-morphemes.md
  sound-symbolism.md
  treasure-hunt-sources.md
scripts/
  naming-roots.py
```

## Usage

Install or copy this folder into an agent skills directory, then invoke it for requests like:

- “Help me name this product.”
- “What should I call this platform?”
- “Rename this feature.”
- “Generate brand name candidates.”
- “Evaluate these names.”

The agent should read `SKILL.md` and follow the methodology before generating candidates.

## Philosophy

Great names do three things:

1. Get attention.
2. Hold attention.
3. Surprise.

Safe names often feel good in meetings and disappear in the market. This skill is designed to push toward the tension zone where names have energy.

## License

MIT License.

# Roots & Morphemes by Concept

Organized by the concepts most commonly needed in naming. Each entry: root, origin, meaning, and example derivatives.

---

## Light / Brightness / Vision

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| lux, luc- | Latin | light | lucid, luminous, Luxe, translucent |
| phot-, phos- | Greek | light | photon, phosphor, phosphorescent |
| clar- | Latin | clear, bright | clarity, declare, Claire |
| lumen, lumin- | Latin | light, lamp | illuminate, luminaire, luminary |
| aur- | Latin | gold, dawn | aurora, auric, aureate |
| sol- | Latin | sun | solar, solstice, soleil (Fr.) |
| heli- | Greek | sun | helio, Helios, heliotrope |
| radi- | Latin | ray, beam | radiant, radiance, radius |
| fulgur- | Latin | lightning, flash | fulgent, effulgent |
| splend- | Latin | shine | splendid, splendor, resplendent |

## Speed / Motion / Flow

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| veloc- | Latin | speed | velocity, veloce |
| celer- | Latin | swift | accelerate, celerity |
| kine-, cine- | Greek | movement | kinetic, kinesis, cinema |
| flux, flu- | Latin | flow | fluent, flux, fluid, affluent |
| curr-, curs- | Latin | run | current, cursor, excursion |
| vol- | Latin | fly | volatile, volant |
| rap- | Latin | seize, swift | rapid, raptor, rapture |
| tach- | Greek | speed | tachyon, tachometer |
| agil- | Latin | nimble | agile, agility |
| drom- | Greek | running, course | hippodrome, syndrome, palindrome |

## Strength / Power / Force

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| fort- | Latin | strong | fortify, forte, fortress |
| dyna- | Greek | power | dynamic, dynamo, dynasty |
| potent- | Latin | powerful | potent, potential, omnipotent |
| rob- | Latin/Germanic | strong | robust, corroborate |
| val- | Latin | strong, worth | valor, valid, prevail |
| krat- | Greek | power, rule | democracy, aristocrat |
| arch- | Greek | chief, first | architect, monarch, archetype |
| magn- | Latin | great | magnitude, magnify, magnificent |
| ferr- | Latin | iron | ferrous, ferric |
| titan- | Greek | giant, powerful | titanic, titanium |

## Small / Precise / Micro

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| micro- | Greek | small | microscope, microbe |
| min- | Latin | small, less | minimal, diminish, minute |
| atom- | Greek | indivisible | atomic, atom |
| punc- | Latin | point | punctual, puncture, pinpoint |
| parv- | Latin | small | parvenu |
| lepto- | Greek | thin, fine | lepton |
| nano- | Greek | dwarf | nanotechnology |
| brev- | Latin | short | brevity, abbreviate, brief |
| acu- | Latin | sharp, precise | acute, acumen, accuracy |
| subtil- | Latin | fine, delicate | subtle, subtlety |

## Transformation / Change / New

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| morph- | Greek | form, shape | metamorphosis, morphology |
| mut- | Latin | change | mutate, mutation, mutable |
| nov- | Latin | new | nova, novel, novice, innovate |
| neo- | Greek | new | neon, neophyte, neo |
| gen- | Greek/Latin | birth, create | genesis, generate, genetic |
| vert-, vers- | Latin | turn | convert, versatile, vertex |
| trans- | Latin | across, beyond | transform, transcend, transit |
| meta- | Greek | beyond, change | metabolism, metaphor |
| proto- | Greek | first | prototype, protocol |
| phas-, phen- | Greek | appear, show | phase, phenomenon, epiphany |

## Connection / Together / Network

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| syn-, sym- | Greek | together | sync, symphony, synergy |
| nexus, nect- | Latin | bind, connect | nexus, connect, annex |
| junct- | Latin | join | junction, conjunction |
| soci- | Latin | companion | social, associate |
| com-, con- | Latin | with, together | combine, converge, commune |
| liga- | Latin | bind | ligature, league, alliance |
| reti- | Latin | net | reticulate, reticular |
| plex- | Latin | weave, fold | complex, multiplex |
| tele- | Greek | far, distant | telescope, telecom |
| inter- | Latin | between | interface, interlink |

## Earth / Nature / Growth

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| terra- | Latin | earth | terrain, terrace, terra |
| geo- | Greek | earth | geology, geometry |
| phyt-, phyto- | Greek | plant | phytoplankton, neophyte |
| bio- | Greek | life | biology, biome, biopsy |
| flor- | Latin | flower, bloom | flora, flourish, Florence |
| silv- | Latin | forest | sylvan, silviculture |
| aqua- | Latin | water | aquatic, aquifer |
| petr- | Greek/Latin | rock, stone | petrify, petroleum, Peter |
| verd- | Latin | green | verdant, verdure |
| arbor- | Latin | tree | arboreal, arboretum |

## Mind / Knowledge / Thought

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| cogn- | Latin | know | cognition, recognize |
| soph- | Greek | wisdom | philosophy, sophisticated |
| ment- | Latin | mind | mental, mentality |
| nous, noo- | Greek | mind, intellect | noosphere, paranoia |
| sci- | Latin | know | science, conscious, prescient |
| gnos- | Greek | know | diagnosis, prognosis, agnostic |
| log-, logi- | Greek | reason, word | logic, logos, analogy |
| cept- | Latin | take, grasp | concept, perception, inception |
| phrn-, phron- | Greek | mind, thought | phronesis |
| doc- | Latin | teach | doctrine, document |

## Air / Flight / Elevation

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| aer- | Greek | air | aerial, aeronautic, aero |
| lev- | Latin | light, lift | levitate, lever, elevate, alleviate |
| alt- | Latin | high | altitude, alto, exalt |
| pter- | Greek | wing | pterodactyl, helicopter |
| vol- | Latin | fly | volant, volatile |
| zephyr- | Greek | west wind | zephyr |
| anem- | Greek | wind | anemone, anemometer |
| plum- | Latin | feather | plume, plumage |
| celest- | Latin | heavenly | celestial |
| ether- | Greek | upper air | ethereal, ether |

## Fire / Energy / Heat

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| pyr- | Greek | fire | pyre, pyro, pyrex |
| ign- | Latin | fire | ignite, ignition |
| therm- | Greek | heat | thermal, thermometer |
| cal- | Latin | heat | calorie, caldera |
| ard- | Latin | burn | ardent, ardor |
| ferv- | Latin | boil, glow | fervent, fervor, effervescent |
| cand- | Latin | glow, white | candle, candid, incandescent |
| flam- | Latin | flame | flame, inflame, flamboyant |
| volt- | Italian | charge | volt, voltage (from Volta) |
| lum- | Latin | light/fire | luminous, luminary |

## Sound / Voice / Harmony

| Root | Origin | Meaning | Derivatives |
|------|--------|---------|-------------|
| son- | Latin | sound | sonic, sonorous, resonance, Sonos |
| phon- | Greek | voice, sound | phonetic, symphony, microphone |
| voc-, vox- | Latin | voice, call | vocal, evoke, vox |
| aud- | Latin | hear | audio, audible, auditorium |
| echo- | Greek | reflected sound | echo |
| ton- | Greek | tone, stretch | tone, tonic, monotone |
| chor- | Greek | dance, chorus | chorus, choreography |
| cant- | Latin | sing | chant, canticle, incantation |
| reson- | Latin | resound | resonate, resonance |
| harmon- | Greek | fitting together | harmony, harmonic |

---

## Quick Reference: Most Productive Roots for Naming

These roots appear most frequently in successful brand names because they combine meaning with strong sound profiles:

**Power roots:** dyna-, magn-, fort-, titan-, arch-
**Innovation roots:** neo-, proto-, gen-, nova-
**Speed roots:** veloc-, celer-, flux-
**Elegance roots:** lux-, aur-, clar-, splend-
**Connection roots:** syn-, nexus-, com-
**Edge roots:** acu-, ferr-, volt-

## Using This File

1. Identify the **ultimate benefit** of the product (Step 3 in the naming process)
2. Find the concept family that matches (light, speed, strength, etc.)
3. Mine roots for raw material — combine, truncate, blend
4. Cross-pollinate: combine a root from one family with a word from another
5. For deeper digs into a specific concept, use the `naming-roots.py` script

# Sound Symbolism & Linguistic Science

The science behind why certain names feel right. Based on decades of cross-language research in sound symbolism, processing fluency, and cognitive linguistics.

## Processing Fluency

The brain's ease of processing a word. Higher fluency = better recall, more positive associations.

### What increases fluency:
- **Familiar phonetic patterns** — sounds that exist in the listener's language
- **CVCV structure** — Consonant-Vowel-Consonant-Vowel (mama, Sonos, Zara)
- **Alliteration** — repeated initial sounds (BlackBerry, Coca-Cola, HubSpot)
- **Rhyme or near-rhyme** — internal sound echoes (SlimFast, YouTube)
- **Short syllable count** — 2-3 syllables is the sweet spot

### The paradox:
Fluency alone = forgettable. You need fluency + surprise.
> "Surprisingly familiar — the brain is a little lazy. It wants easy processing. But then: oh, there's something interesting here."

## Sound-Letter Associations

### Speed & Energy
| Sound | Speed Rating | Examples |
|-------|-------------|----------|
| Z | Very fast | Zoom, Zara, Amazon |
| X | Fast, innovative | SpaceX, Lexus, Xbox |
| K | Fast, precise | Kayak, Kodak, Nike |
| P | Quick, punchy | Pentium, Pepsi, Ping |
| D | Dynamic, direct | Dell, Discord, Dash |
| T | Tight, technical | Tesla, Twitter, TikTok |

### Reliability & Warmth
| Sound | Association | Examples |
|-------|------------|----------|
| B | Bold, reliable, warm | Blackberry, Bose, Boeing |
| M | Soft, maternal, comforting | Method, Miele, Muji |
| N | Smooth, natural | Napa, Nissan, Nokia |
| L | Flowing, elegant | Lexus, Lyft, Lululemon |
| S | Smooth, sophisticated | Sonos, Sephora, Slack |

### Innovation & Edge
| Sound | Association | Examples |
|-------|------------|----------|
| X | Unknown, cutting edge | Xero, Xerox, SpaceX |
| Q | Quirky, quality, quantum | Quora, Qloo, Quest |
| V | Velocity, vitality | Vercel, Visa, Vivid |

## The CVCV Principle

Children's first words are universally CVCV: mama, dada, papa, nana.

This pattern is hardwired for easy processing. Names that follow it benefit:
- **Sonos** — CVCVC (close enough)
- **Toro** — CVCV (perfect)
- **Zara** — CVCV (perfect)
- **SoHo** — CVCV (perfect)
- **Roku** — CVCV (perfect)

Not every name needs to be CVCV, but understanding the pattern helps explain why some names feel instantly "right."

## Cross-Language Considerations

### Universally recognized sounds/words:
- "Black" — recognized across many languages
- Color words generally travel well
- Hard consonants (K, T, P) are universal
- Vowels vary significantly across languages

### Danger zones:
- Words that mean something negative in major languages
- Sounds that don't exist in target markets (e.g., "th" in many Asian languages)
- Names that can't be written in non-Latin scripts
- Always check: Spanish, Mandarin, Japanese, German, French, Portuguese, Arabic at minimum

## The Surprisingly Familiar Formula

The best names combine:

1. **A familiar element** (real word, recognizable root, common sound pattern)
2. **An unexpected context** (that word doesn't belong in this category)

| Name | Familiar Element | Unexpected Context |
|------|-----------------|-------------------|
| Blackberry | A fruit | On a tech device |
| Impossible | Common adjective | On a burger |
| Swiffer | Sounds like "swift" | For floor cleaning |
| Azure | A color word | For cloud computing |
| Feather | A nature word | For a fiber supplement |
| Windsurf | A sport | For a code editor |

> "Your current competitors would NEVER have the courage to put Blackberry on a device."

That "never" is the test. If a competitor would use this name, it's not surprising enough.

## Placek's Scoring Framework

For each name candidate, evaluate:

```
ORIGINAL       [1-10]  Unlike anything in this category?
FLUENT         [1-10]  Easy to say, spell, and grasp?
SURPRISING     [1-10]  Would competitors never dare use this?
COMPOUNDS      [1-10]  Gets stronger over time?
BELIEVABLE     [1-10]  In context, does it feel real?
POLARIZING     [Y/N]   Does it split the room? (Good if yes)
```

No sevens. Sevens are for wimps.

# Treasure Hunt Sources

Where to dig for name candidates. Each source is a database to mine — not for direct answers, but for raw material that sparks connections.

## Linguistic Databases

### Latin & Greek Roots
- Latin root dictionaries (online: latin-dictionary.net, Perseus Digital Library)
- Greek word units — prefixes, suffixes, root words
- Medical/scientific Latin (often yields surprisingly elegant words)
- Example: "Pentium" — from Greek "pente" (five), for the 5th-generation processor

### Mythology
- Roman mythology (gods, places, concepts)
- Greek mythology (same, but different flavor)
- Norse mythology (strong, sharp sounds)
- Example: Look up gods/figures associated with your product's ultimate benefit

### Periodic Table (Enriched)
- Don't just scan element names — read about their properties, discovery stories, etymologies
- Element names often come from places, people, or properties
- The enriched version: articles, histories, and associations around each element

### Foreign Languages
- Romance languages (Italian, French, Spanish, Portuguese) for elegance
- Germanic languages for strength/reliability
- Japanese for precision/minimalism
- Look for words that are surprisingly familiar across languages
- Example: "Azure" — understood across many languages, signals sky/cloud

## Conceptual Databases

### Cross-Domain Collections
Build or search for collections organized by CONCEPT, not product:
- "Small things" — could serve both a processor and a car
- "Fast things" — serves tech, sports, automotive
- "Light things" — serves food, tech, aerospace
- "Reliable things" — serves finance, infrastructure, health

The synchronicity principle: names from unrelated domains create the most surprising connections.

### The Unrelated Magazine Exercise
Spend 30 minutes with 2 magazines you've NEVER read before.
- Look for connections to your project
- Don't evaluate — just note anything that sparks
- ~30% of the time this produces a new perspective or insight
- This is deliberately inefficient. That's the point.

### Book Titles & Chapter Headings
- Scan bookstore shelves (physically or online)
- Children's books are especially rich (simple, vivid, memorable language)
- Poetry collections (compressed language, unusual word combinations)

## Sound & Structure Resources

### Sound Symbolism Research
- Phonesthemes: sound clusters that carry meaning (gl- = light/vision: glow, gleam, glitter)
- Bouba/Kiki effect: rounded sounds feel soft, sharp sounds feel angular
- Speed sounds: Z, X, P, D, K (fast); S, M, L, N (slow/smooth)
- Reliability sounds: B, D, strong consonants

### CVCV Pattern Words
Consonant-Vowel-Consonant-Vowel — the first pattern children learn:
- mama, dada, Sonos, Toro, Napa, Soho, Zara
- These are inherently easy to process and remember

### Power Letters
| Letter | Association | Examples |
|--------|-------------|----------|
| K | Energy, precision, sharpness | Kodak, Nike, Kayak |
| X | Innovation, unknown, edge | SpaceX, Lexus, Xerox |
| Z | Speed, electricity, cutting edge | Zoom, Zara, Amazon |
| P | Pop, power, precision | Pentium, Pepsi, PayPal |
| B | Reliability, boldness, warmth | Blackberry, Boeing, Bose |

## Process Notes

### Quantity Targets
- Raw candidates from all sources: ~2,000
- After first linguistic filter: ~200
- After trademark preliminary: ~50
- Proof of concept round: ~15
- Final presentation: 5-7

### Two-Person Teams
- Each team gets a different lens (direct, adjacent, distant)
- No group brainstorming — peer pressure kills divergent thinking
- Permission to fail is explicit: "If we're not failing, we're not doing our job"
- Use "I wish we could..." instead of "That won't work" (converts criticism into problem-solving)

### AI as Treasure Hunt Tool
- Use AI to explore domains, not to generate names directly
- Ask questions about the product's world, not "give me 10 names"
- AI is good at: etymology, cross-language checks, finding related concepts
- AI is bad at: the surprise factor, the polarization test, believability judgment
- The human advantage: judging which names are RIGHT, not just good

#!/usr/bin/env python3
"""
naming-roots.py — Etymology and root word explorer for naming projects.

Queries Wiktionary API for etymologies, roots, and derived terms.
Falls back to web scraping Etymonline when Wiktionary lacks data.

Usage:
    naming-roots.py <concept> [<concept>...]
    naming-roots.py --root <root>
    naming-roots.py --blend <word1> <word2>

Examples:
    naming-roots.py light speed transform
    naming-roots.py --root "lux"
    naming-roots.py --blend "swift" "feather"
"""

import sys
import json
import urllib.request
import urllib.parse
import re
from typing import Optional


def fetch_json(url: str, timeout: int = 10) -> Optional[dict]:
    """Fetch JSON from a URL."""
    try:
        req = urllib.request.Request(url, headers={"User-Agent": "NamingRoots/1.0"})
        with urllib.request.urlopen(req, timeout=timeout) as resp:
            return json.loads(resp.read().decode())
    except Exception:
        return None


def wiktionary_lookup(word: str) -> dict:
    """Look up a word on Wiktionary for etymology and related terms."""
    url = f"https://en.wiktionary.org/api/rest_v1/page/definition/{urllib.parse.quote(word)}"
    data = fetch_json(url)
    result = {"word": word, "definitions": [], "etymology": None}

    if not data:
        return result

    for lang_section in data.get("en", []):
        part = lang_section.get("partOfSpeech", "")
        for defn in lang_section.get("definitions", []):
            text = re.sub(r"<[^>]+>", "", defn.get("definition", ""))
            if text:
                result["definitions"].append({"pos": part, "text": text[:200]})

    # Try to get etymology from the HTML page
    html_url = f"https://en.wiktionary.org/w/api.php?action=parse&page={urllib.parse.quote(word)}&prop=wikitext&format=json"
    html_data = fetch_json(html_url)
    if html_data:
        wikitext = html_data.get("parse", {}).get("wikitext", {}).get("*", "")
        etym_match = re.search(r"===Etymology===\n(.*?)(?=\n===|\n\[\[Category|\Z)", wikitext, re.DOTALL)
        if etym_match:
            etym = etym_match.group(1).strip()
            # Clean up wiki markup
            etym = re.sub(r"\{\{[^}]*\}\}", "", etym)
            etym = re.sub(r"\[\[([^\]|]*\|)?([^\]]*)\]\]", r"\2", etym)
            etym = etym.strip()
            if etym:
                result["etymology"] = etym[:500]

    return result


def etymonline_lookup(word: str) -> Optional[str]:
    """Scrape Etymonline for a word's etymology."""
    url = f"https://www.etymonline.com/word/{urllib.parse.quote(word)}"
    try:
        req = urllib.request.Request(url, headers={"User-Agent": "NamingRoots/1.0"})
        with urllib.request.urlopen(req, timeout=10) as resp:
            html = resp.read().decode()
        # Extract the etymology text from the page
        match = re.search(r'class="word--[^"]*__defination[^"]*"[^>]*>(.*?)</section>', html, re.DOTALL)
        if match:
            text = re.sub(r"<[^>]+>", "", match.group(1))
            text = re.sub(r"\s+", " ", text).strip()
            return text[:600] if text else None
    except Exception:
        pass
    return None


def find_related_words(concept: str) -> list:
    """Use Wiktionary category/search to find related words."""
    url = f"https://en.wiktionary.org/w/api.php?action=opensearch&search={urllib.parse.quote(concept)}&limit=20&format=json"
    data = fetch_json(url)
    if data and len(data) > 1:
        return data[1][:15]
    return []


def generate_blends(word1: str, word2: str) -> list:
    """Generate portmanteau/blend candidates from two words."""
    blends = []

    # Front of word1 + back of word2
    for i in range(2, len(word1)):
        for j in range(0, len(word2) - 1):
            blend = word1[:i] + word2[j:]
            if 4 <= len(blend) <= 10:
                blends.append(blend)

    # Front of word2 + back of word1
    for i in range(2, len(word2)):
        for j in range(0, len(word1) - 1):
            blend = word2[:i] + word1[j:]
            if 4 <= len(blend) <= 10:
                blends.append(blend)

    # Deduplicate and filter
    seen = set()
    unique = []
    for b in blends:
        b_lower = b.lower()
        if b_lower not in seen and b_lower != word1.lower() and b_lower != word2.lower():
            seen.add(b_lower)
            unique.append(b)

    # Score by pronounceability (vowel/consonant alternation)
    def score(w):
        vowels = set("aeiou")
        transitions = sum(1 for i in range(len(w)-1) if (w[i].lower() in vowels) != (w[i+1].lower() in vowels))
        ratio = transitions / max(len(w) - 1, 1)
        return ratio

    unique.sort(key=score, reverse=True)
    return unique[:25]


def explore_concept(concept: str):
    """Full exploration of a concept for naming purposes."""
    print(f"\n{'='*60}")
    print(f"  CONCEPT: {concept.upper()}")
    print(f"{'='*60}")

    # 1. Direct lookup
    print(f"\n--- Wiktionary: '{concept}' ---")
    wiki = wiktionary_lookup(concept)
    if wiki["etymology"]:
        print(f"Etymology: {wiki['etymology']}")
    if wiki["definitions"]:
        for d in wiki["definitions"][:3]:
            print(f"  [{d['pos']}] {d['text']}")

    # 2. Etymonline
    print(f"\n--- Etymonline: '{concept}' ---")
    etym = etymonline_lookup(concept)
    if etym:
        print(etym)
    else:
        print("(no entry found)")

    # 3. Related words
    print(f"\n--- Related words ---")
    related = find_related_words(concept)
    if related:
        print(", ".join(related))
    else:
        print("(none found)")

    # 4. Common roots for this concept
    print(f"\n--- Suggested roots to explore ---")
    roots = find_related_words(concept + " root")
    latin = find_related_words(concept + " Latin")
    greek = find_related_words(concept + " Greek")
    all_suggestions = list(set(roots + latin + greek))[:15]
    if all_suggestions:
        print(", ".join(all_suggestions))
    else:
        print("(use the roots-and-morphemes.md reference file)")


def explore_root(root: str):
    """Deep dive on a specific root."""
    print(f"\n{'='*60}")
    print(f"  ROOT: {root}")
    print(f"{'='*60}")

    wiki = wiktionary_lookup(root)
    if wiki["etymology"]:
        print(f"\nEtymology: {wiki['etymology']}")
    if wiki["definitions"]:
        print("\nDefinitions:")
        for d in wiki["definitions"][:5]:
            print(f"  [{d['pos']}] {d['text']}")

    etym = etymonline_lookup(root)
    if etym:
        print(f"\nEtymonline: {etym}")

    related = find_related_words(root)
    if related:
        print(f"\nRelated/derived: {', '.join(related)}")


def blend_words(word1: str, word2: str):
    """Generate and display blend candidates."""
    print(f"\n{'='*60}")
    print(f"  BLENDING: {word1} + {word2}")
    print(f"{'='*60}")

    blends = generate_blends(word1, word2)
    if blends:
        print(f"\nTop blend candidates (sorted by pronounceability):\n")
        for i, b in enumerate(blends, 1):
            print(f"  {i:2d}. {b}")
    else:
        print("(no viable blends found)")


def main():
    if len(sys.argv) < 2:
        print(__doc__)
        sys.exit(1)

    if sys.argv[1] == "--root" and len(sys.argv) >= 3:
        explore_root(sys.argv[2])
    elif sys.argv[1] == "--blend" and len(sys.argv) >= 4:
        blend_words(sys.argv[2], sys.argv[3])
    else:
        concepts = sys.argv[1:] if sys.argv[1] != "--" else sys.argv[2:]
        for concept in concepts:
            explore_concept(concept)


if __name__ == "__main__":
    main()

---
name: naming
description: Name companies, products, features, and projects using David Placek's professional naming methodology. Use when asked to "name my company", "come up with a name", "what should I call this", "brand name ideas", "rename this", "help me name", or any naming/branding challenge. Applies the Lexicon Branding framework — the firm behind Blackberry, Swiffer, Impossible Burger, Pentium, Sonos, Febreze, and Windsurf.
---

# /naming — The Placek Method

Professional naming methodology extracted from David Placek (Lexicon Branding), the mind behind Blackberry, Swiffer, Impossible Burger, Intel Pentium, Sonos, Febreze, Windsurf, SlimFast, and Microsoft Azure.

**This is not brainstorming.** This is a structured treasure hunt with linguistics, cognitive science, and creative curiosity.

## Core Principles

### The Three Jobs of a Name
Every great name does three things:
1. **Gets attention** — stands out in a crowded category
2. **Holds attention** — is "processing fluent" (easy to say, something to grasp)
3. **Surprises** — is unexpected, NOT comfortable, NOT safe

> "The right name creates asymmetric advantage. Impossible does that. Swiffer does that."

### The Comfort Trap
Most names fail because they're safe. The **tension zone** — where half the team hates it — is where the energy lives.

- **Invisible zone** (bottom): High consensus, low distinctiveness. Safe. Forgettable. "ProMop."
- **Tension zone** (middle): Polarizing. High energy. "Swiffer." "Blackberry." "Impossible."

> "If it's polarizing for your organization, that's GOOD. It means it has energy inside."

### Compound Value
A name is the highest-frequency leverage point in a brand. It compounds over time. The difference between an okay name and the right name grows every day. Get it right early.

## The Process

### Step 0: Product / Category Intake (MANDATORY FIRST)
Before the Comfort Trap Interview or any name generation, learn what is being named. Do not ask category-relative questions until the category exists.

Interview the user and/or read the supplied context doc to understand the topics below, but **do not fire them all at the user at once**. Run this as a real back-and-forth: ask **one question at a time**, wait for the answer, then decide the next best question. If a context doc already answers something, do not ask it again.

Intake topics to cover gradually:

1. **What is the product/platform?** What does it actually do?
2. **What are the pieces?** Is this one umbrella name, multiple product/module names, or both?
3. **Who uses it and who buys it?** End users, economic buyers, influencers, internal champions.
4. **What category does it compete in?** What would buyers compare it against?
5. **What is the product definition still unsure about?** Identify fuzzy boundaries before naming them.
6. **What must the name accomplish strategically?** Signal trust, speed, intelligence, creativity, power, ease, rebellion, etc.

Only after this intake should you run the Comfort Trap Interview, also **one question at a time**. If the product/category is unclear, pause and ask for the single most important missing piece rather than performing naming theater in the void.

### Step 0.5: The Comfort Trap Interview (MANDATORY)
Before generating a single name, have this conversation with the user. The goal is to surface their unconscious bias toward safety and set expectations.

**Ask these questions in order, one at a time. Wait for the user's response before asking the next question.**

1. **"Give me an example of a name in your space that you think is good."**
   → Reveals their gravitational pull. If they name something descriptive and safe, that's the trap to work against.

2. **"If your biggest competitor launched with a name that felt totally unexpected — a fruit, an animal, an emotion — what would your gut reaction be?"**
   → Tests whether they'd dismiss the very thing that would work. "That's weird" from a founder usually means "that would get attention."

3. **"Are you naming this for YOU or for the MARKET?"**
   → Founders optimize for names they feel comfortable saying in board meetings. That's not the same as what cuts through to customers.

4. **"What name would you NEVER use?"**
   → The rejection list is more revealing than the wish list. The names they'd never use may be exactly the tension zone candidates.

5. **Set the ground rules:**
   > "Every platinum name on David Placek's wall — Blackberry, Swiffer, Pentium, Impossible — was rejected by the client at first. If every name I show you feels comfortable, I've failed you. If something makes you flinch — that's the one we investigate."

Use their answers to calibrate the entire process. If they're already comfort-trapped (all examples are descriptive/safe), push harder toward the surprising end. If they're already open to bold names, you can explore the full spectrum.

### Step 1: Landscape Audit
Map the category. What does everyone else sound like?

- List all competitors' names
- Identify the naming conventions (descriptive? techy? nature words?)
- **First hypothesis: "We're not going there."** You now know what to avoid.

> Ready Mop vs Swiffer. Fiber One vs Feather. ProMop vs Swiffer. The comfortable name always loses.

### Step 2: Deep Product Understanding
Go beyond the feature list:

1. **What do you have?** (the product, its actual properties)
2. **How do you define winning?** (ask 5 people — every answer will differ)
3. **What do you have to win?** (competitive advantages)
4. **What do you need to win?** (market dynamics, positioning gaps)
5. **What do you need to SAY?** (the core message the name must carry)

### Step 3: Find the Ultimate Benefit
Climb the benefit ladder. Don't name the feature — name what the customer FEELS.

**Example — Fiber product:**
- Feature level: "Has fiber" → Fiber One, BenFiber (boring, invisible)
- Benefit level: "Gut health" (better but clinical)
- Ultimate benefit: "You feel LIGHTER" → **Feather** (surprising, processing fluent, original in category)

> "The Swiffer principle: It's not a mop. Don't name it like a mop."

### Step 4: Treasure Hunt — Generate 2,000+ Candidates
**Quantity leads to quality.** 50-100 names is where people get stuck. You need ~2,000.

Work in **small teams of 1-2 people** (never brainstorming groups — peer pressure kills creativity).

Run **three parallel tracks** with different lenses:
1. **Direct lens** — knows everything about the product
2. **Adjacent lens** — product + an added twist (e.g., fiber + energy)
3. **Distant lens** — completely different domain (e.g., athletic performance, aviation)

**Treasure hunting databases** (see `references/treasure-hunt-sources.md`):
- Latin/Greek roots and word units → use `references/roots-and-morphemes.md` for instant lookup, or `scripts/naming-roots.py <concept>` for deeper etymology digs
- Mythology (Roman, Greek, Norse) → web search at runtime
- Periodic table (enriched with contextual articles) → web search at runtime
- Sound symbolism research → see `references/sound-symbolism.md`
- Foreign language words → `scripts/naming-roots.py --root <word>` or web search
- Cross-domain databases ("small things", "fast things", "light things") → `roots-and-morphemes.md` organized by concept
- Word blending → `scripts/naming-roots.py --blend <word1> <word2>` generates portmanteau candidates
- Magazines/books you've never read (synchronicity — connecting seemingly irrelevant things)

> "The fundamental layer of creativity is connecting seemingly irrelevant things."

### Step 5: Apply Linguistic Science
Filter candidates through cognitive science. See `references/sound-symbolism.md`.

**Key filters:**
- **Processing fluency** — can the brain process it easily? Is there something familiar AND surprising?
- **CVCV pattern** — Consonant-Vowel-Consonant-Vowel is how children learn language (mama, dada, Sonos, Toro)
- **Power letters** — K, P, B, Z, X carry energy and speed
- **Sound symbolism** — letters have inherent associations (speed, reliability, softness, strength)
- **Alliteration** — increases memorability (HubSpot, BlackBerry, Coca-Cola)

### Step 6: Believability Test (Proof of Concept)
Don't present names on a spreadsheet. Put them in context:

- A Wall Street Journal headline
- On a product label next to competitors
- In a tagline: "[Name]: the lightweight fiber you need"
- On a billboard, a t-shirt, a conference badge

**The one rule:** Is it BELIEVABLE? In less than one second, does the viewer lean toward "I think I believe this"?

### Step 7: Evaluate Against the Framework
For each finalist, score on:

| Criterion | Question |
|-----------|----------|
| **Original** | Is it unlike anything else in this category? |
| **Processing fluent** | Can someone say it, spell it, and grasp something about it? |
| **Surprising** | Would competitors NEVER have the courage to use this? |
| **Compounds** | Will this name get stronger over time, not weaker? |
| **Searchable** | Can you find it on Google? Is the domain available or acquirable? |
| **Cross-market** | Does it work across languages? Any negative connotations? |
| **Believable** | In context (headline, label, ad), does it feel real? |

### The Approximate Thinking Line
When evaluating with teams, use this mental model:

```
BIZARRE/ABSURD ←——— APPROXIMATE THINKING ———→ SAFE/WORKABLE
```

Start from the bizarre end. Move toward approximate. **Stop there.** Evaluate approximate ideas — not fully baked, not safe. This is where the best names live.

## Anti-Patterns (What NOT to Do)

1. **Don't describe the product** — ProMop, ReadyMop, Fiber One. You'll be invisible.
2. **Don't brainstorm in groups** — Peer pressure, evaluation cascade, slow grind. Work solo or in pairs.
3. **Don't stop at 100 names** — That's where everyone gets stuck. Push to 2,000.
4. **Don't test names by asking "what do you think this is?"** — Comfortable names win that test and lose in the market.
5. **Don't optimize for consensus** — Consensus = invisible zone. Polarization = energy.
6. **Don't fear losing equity when renaming** — Evidence shows companies never lose momentum from a name change if the launch tells a story. Kodium → Windsurf. The friction of a bad name costs more than the change.

## Quick-Fire Naming (5-Day Process)

When full methodology isn't practical:

**Day 1:** Landscape audit + product deep-dive + ultimate benefit identification
**Day 2:** Solo treasure hunt across 5+ source databases. Generate 500+ raw candidates.
**Day 3:** Apply linguistic filters. Narrow to ~50. Do the "unrelated magazine" exercise — spend 30 minutes with two magazines you've never read, looking for connections.
**Day 4:** Proof of concept — put top 15 in headlines, labels, taglines. Believability test.
**Day 5:** Trademark preliminary search. Domain check. Final presentation of 5-7 names with context.

## Recommended Reading
- *Ogilvy on Advertising* — David Ogilvy (principles of great communication)
- *Confessions of an Advertising Man* — David Ogilvy
- *Leonardo da Vinci* — Walter Isaacson (creative persistence and curiosity)
- *Playing to Win* — Roger Martin & A.G. Lafley (strategic framing)
- *A New Way to Think* — Roger Martin

## When to Use This Skill

- Naming a new company, product, feature, or project
- Renaming something that has a "ProMop" problem
- Evaluating proposed names against a framework
- Generating name candidates for any context
- Understanding why a current name isn't working