language-evolution
架空の世界における言語の進化システムを設計します。言語系統や方言、言語の歴史を構築したい場合、あるいは言語が文化・歴史的な発展を反映すべき場面で活用してください。
description の原文を見る
Design evolving language systems for fictional worlds. Use when creating language families, dialects, linguistic history, or when language should reflect cultural and historical development.
SKILL.md 本文
Language Evolution: Linguistic Development Skill
You help writers create realistic language systems that evolve over time and reflect cultural history. This goes beyond conlang phonology to address how languages change, branch, and interact across generations and geographies.
Core Principles
- Historical Continuity: Languages evolve from previous forms rather than appearing fully formed
- Contact Modification: Languages change through interaction with other languages
- Functional Adaptation: Language structures evolve to serve communication needs
- Cultural Reflection: Languages encode values, environment, and practices of speakers
- Cognitive Constraints: Development is shaped by human cognitive limitations
- Register Variation: Languages develop specialized forms for different contexts
- Innovation-Conservation Balance: Languages contain both innovative and conservative elements
- Geographic Divergence: Physical separation leads to linguistic divergence over time
- Sociolinguistic Stratification: Language varies across social groups
- Writing System Independence: Spoken and written forms evolve semi-independently
Parameter Categories
1. Environmental Parameters
| Parameter | What It Affects |
|---|---|
| Geographic Distribution | Mountain ranges, rivers affecting spread |
| Climate Influence | Weather and seasonal vocabulary |
| Resource Availability | Local materials in terminology |
| Fauna and Flora | Taxonomic complexity for important species |
| Topographical Marking | Landscape feature naming patterns |
2. Cultural-Historical Parameters
| Parameter | What It Affects |
|---|---|
| Migration Patterns | Population movements creating contact |
| Conquest History | Dominant-subordinate language relationships |
| Trade Networks | Commercial contact creating exchange |
| Technological Development | New terminology requirements |
| Religious Traditions | Abstract concepts and sacred language |
3. Sociolinguistic Parameters
| Parameter | What It Affects |
|---|---|
| Social Stratification | Class-based language variation |
| Occupational Specialization | Professional jargons |
| Gender Differentiation | Gender-based language patterns |
| Age Grading | Generational change markers |
| Group Identity Marking | In-group terminology and pronunciation |
4. Communication Context Parameters
| Parameter | What It Affects |
|---|---|
| Formality Levels | Situational appropriateness markers |
| Medium Adaptation | Spoken vs. written vs. digital |
| Specialist Discourse | Technical, legal, scientific evolution |
| Artistic Expression | Poetic, narrative, performance forms |
| Privacy/Secrecy | Coded communication and euphemisms |
Language Typologies
Morphological Types
| Type | Characteristics | Real Examples |
|---|---|---|
| Isolating | Minimal word modification | Mandarin Chinese |
| Agglutinative | Clear morpheme boundaries | Turkish, Japanese |
| Fusional | Multiple meanings in single morphemes | Latin, Russian |
| Polysynthetic | Many morphemes per word | Inuktitut, Mohawk |
Word Order Types
| Type | Pattern | Examples |
|---|---|---|
| SVO | Subject-Verb-Object | English, Mandarin |
| SOV | Subject-Object-Verb | Japanese, Turkish |
| VSO | Verb-Subject-Object | Irish, Classical Arabic |
| VOS | Verb-Object-Subject | Malagasy |
| OVS | Object-Verb-Subject | Hixkaryana |
| OSV | Object-Subject-Verb | Rare |
Writing System Types
| Type | How It Works | Examples |
|---|---|---|
| Logographic | Character per word/morpheme | Chinese |
| Syllabic | Character per syllable | Japanese kana |
| Alphabetic | Character per phoneme | Latin, Cyrillic |
| Abjad | Consonants primarily | Arabic, Hebrew |
| Abugida | Consonant-vowel units | Devanagari |
| Featural | Characters represent features | Korean Hangul |
Language Evolution Mechanisms
Sound Change Types
| Type | Description |
|---|---|
| Lenition | Weakening of consonants |
| Fortition | Strengthening of consonants |
| Vowel Shift | Systematic vowel changes |
| Palatalization | Consonants shift toward palate |
| Assimilation | Sounds become more similar |
| Metathesis | Sound order swaps |
Grammatical Evolution
| Type | Description |
|---|---|
| Grammaticalization | Lexical words become grammatical |
| Analogical Leveling | Irregular forms become regular |
| Case System Simplification | Loss of case distinctions |
| Tense/Aspect Development | New temporal distinctions |
| Evidentiality Emergence | Source marking becomes grammatical |
Contact Effects
| Type | Description |
|---|---|
| Lexical Borrowing | Vocabulary adoption (most common) |
| Phonological Influence | Sound system adjustments |
| Syntactic Convergence | Sentence structure alignment |
| Morphological Simplification | Complexity reduction in contact |
| Calquing | Loan translation with native words |
| Code-Switching | Alternation between languages |
Language Family Construction
Step 1: Proto-Language Design
- Create core vocabulary (200-500 words)
- Establish basic phoneme inventory
- Define grammatical skeleton
- Set morphological type
Step 2: Sound Change Rules
- Define systematic sound shifts
- Apply changes to create daughter languages
- Track which changes apply where
- Create regular correspondences
Step 3: Grammatical Divergence
- Develop distinct innovations per branch
- Create unique grammatical features
- Track loss and gain of categories
- Design independent evolution paths
Step 4: Vocabulary Divergence
- Track cognate relationships
- Add unique vocabulary per branch
- Create borrowings from contact
- Develop semantic shifts
Step 5: Contact Zone Development
- Map where languages meet
- Create contact effects
- Develop pidgins/creoles if appropriate
- Design bilingual phenomena
Common Evolution Sequences
Tonal Development
- Consonant distinctions lost → Pitch compensates → Tones stabilize
Case System Simplification
- Full case → Reduced case → Prepositions → Fixed word order
Creolization
- Pidgin → Expanded pidgin → Creole with native speakers
Dialect to Language
- Single language → Regional varieties → Political division → "Separate languages"
Setting-Specific Adaptations
Fantasy Settings
- Elven Language Family: Ancient, conservative, prestige
- Dwarven Isolation: Mountain-separated dialects
- Human Diversity: Rapid change and adaptation
- Magical Terminology: Specialized arcane vocabulary
- Dead Language Remnants: Ritual preservation
Science Fiction Settings
- Post-Earth Divergence: Colony isolation effects
- Alien-Human Pidgins: Contact language development
- Universal Translator Implications: Technology effects
- Digital-Augmented Communication: Tech-language interface
- Xenolinguistic Principles: Non-human cognition
Post-Apocalyptic Settings
- Linguistic Fragmentation: Isolation creating new dialects
- Technological Vocabulary Loss: Terms for lost tech
- Specialized Jargon: New environmental challenges
- Writing System Degradation: Literacy decline effects
- Pre-Collapse Remnants: Preserved texts, misunderstandings
Sociolinguistic Variation
Register Levels
| Register | Context | Features |
|---|---|---|
| Frozen | Ceremonies, oaths | Fixed phrases, archaic forms |
| Formal | Official, professional | Complete sentences, technical |
| Consultative | Teacher-student, expert-client | Standard grammar |
| Casual | Friends, family | Slang, ellipsis |
| Intimate | Close relationships | Private vocabulary |
Dialect Markers
| Type | What Varies |
|---|---|
| Phonological | Pronunciation differences |
| Lexical | Vocabulary differences |
| Grammatical | Structure differences |
| Pragmatic | Usage differences |
Implementation Checklist
- Define language family relationships
- Create proto-language skeleton
- Design sound change rules
- Develop grammatical divergence
- Map sociolinguistic variation
- Create writing system (if any)
- Design contact zone effects
- Build register variation
- Document sample texts
- Create naming conventions integration
Case Study Examples
Tolkien's Languages
- Proto-Eldarin as common ancestor
- Quenya: conservative, prestige (Latin analog)
- Sindarin: evolved, everyday (Romance analog)
- Systematic sound changes documented
- Cultural-linguistic integration
Klingon
- Distinctive phonology matching warrior culture
- Grammar reflecting cultural values
- Vocabulary emphasizing important domains
- Writing system matching technology level
Valyrian (Game of Thrones)
- High Valyrian as classical, learned language
- Daughter languages showing realistic divergence
- Contact effects with other languages
Output Persistence
Output Discovery
- Check for
context/output-config.mdin the project - If found, look for this skill's entry
- If not found, ask user: "Where should I save language evolution work?"
- Suggest:
worldbuilding/languages/orexplorations/worldbuilding/
Primary Output
- Language family tree - Proto-language and daughter branches
- Sound change rules - Systematic transformations per branch
- Grammatical divergence - How branches differ structurally
- Contact zone effects - Borrowings, pidgins, convergence
- Sociolinguistic variation - Registers, dialects, markers
File Naming
Pattern: {language-family}-evolution-{date}.md
Verification (Oracle)
What This Skill Can Verify
- Sound change consistency - Do rules apply systematically? (High confidence)
- Typological plausibility - Does combination of features exist in real languages? (Medium confidence)
- Evolution logic - Do changes follow from contact/isolation patterns? (High confidence)
What Requires Human Judgment
- Aesthetics - Does the language sound right for the culture?
- Story fit - Does linguistic variation serve narrative?
- Reader accessibility - Will readers parse invented words?
Oracle Limitations
- Cannot assess whether language feels "right" for fictional culture
- Cannot predict reader pronunciation assumptions
Feedback Loop
Session Persistence
- Output location: See
context/output-config.md - What to save: Family tree, sound changes, grammatical features, contact effects
- Naming pattern:
{language-family}-evolution-{date}.md
Cross-Session Learning
- Check for prior language work in this world
- Ensure new languages maintain family consistency
- Failed sound changes inform anti-patterns
Design Constraints
This Skill Assumes
- Setting has languages that evolved (not created ex nihilo)
- Writer wants historical depth, not just vocabulary
- Some linguistic diversity exists
This Skill Does Not Handle
- Detailed phonology - Route to: conlang
- Cultural texture - Route to: memetic-depth
- Generational society change - Route to: multi-order-evolution
- Naming conventions - Route to: character-naming
Degradation Signals
- English grammar with substituted words (relexification)
- Languages too regular without exceptions
- No sociolinguistic variation within languages
Reasoning Requirements
Standard Reasoning
- Single sound change application
- Basic grammatical divergence
- Simple dialect variation
Extended Reasoning (ultrathink)
- Full language family design - [Why: sound changes compound across branches]
- Contact zone synthesis - [Why: multiple languages interacting]
- Deep historical development - [Why: tracing evolution across centuries]
Trigger phrases: "design the language family", "how did these languages diverge", "linguistic history"
Execution Strategy
Sequential (Default)
- Proto-language before daughter languages
- Sound changes before applying to vocabulary
- Family structure before contact effects
Parallelizable
- Designing independent language branches
- Researching different linguistic analogs
Subagent Candidates
| Task | Agent Type | When to Spawn |
|---|---|---|
| Linguistic research | general-purpose | When modeling on real language families |
| Conlang phonology | general-purpose | When needing detailed sound inventory |
Context Management
Approximate Token Footprint
- Skill base: ~3k tokens (parameters + mechanisms)
- With typologies: ~4k tokens
- With case studies: ~5k tokens
Context Optimization
- Focus on relevant evolution mechanisms
- Typologies are reference, load on-demand
- Case studies optional examples
When Context Gets Tight
- Prioritize: Current evolution mechanism, active family branch
- Defer: Full typology tables, all mechanisms not in use
- Drop: Case studies, setting-specific adaptations
Anti-Patterns
1. Relexification
Pattern: Creating "alien language" by substituting words into English grammar and syntax—"Klaatu barada nikto" as sentence structure. Why it fails: Language families don't work this way. Different languages have different grammatical structures, word orders, and morphological patterns. English-with-different-words feels fake. Fix: Choose a typological profile different from English. An SOV language with agglutinative morphology will feel genuinely foreign even with limited vocabulary.
2. Perfect Regularity
Pattern: Languages with no exceptions, no irregular verbs, no spelling inconsistencies—logically constructed rather than evolved. Why it fails: Real languages accumulate irregularities through history. The most common words resist change, preserving older forms. Constructed perfection signals artificial origin. Fix: Add irregularity to high-frequency elements. "To be" equivalents should be irregular. Common plurals should have exceptions. Spelling should preserve historical pronunciations.
3. Frozen Languages
Pattern: Languages unchanged for millennia, spoken identically by ancient elves and their modern descendants. Why it fails: All spoken languages change. Geographic separation creates dialects. Prestige languages like Latin fossilize as literary forms while spoken vernacular evolves. Fix: Create at least archaic and modern registers. Show dialect variation across regions. Have characters note "old-fashioned" speech patterns.
4. Contact Without Effect
Pattern: Languages existing side by side for centuries without borrowing, convergence, or pidginization. Why it fails: Language contact always produces change. Trade brings vocabulary. Conquest brings grammatical influence. Bilingualism creates code-switching patterns. Fix: Map where languages meet. Identify domains where borrowing occurs (technology, trade goods, governance). Create contact phenomena appropriate to relationship type.
5. Monolingual Societies
Pattern: Everyone in a kingdom speaking exactly one language with no regional variation, no professional jargon, no class markers. Why it fails: Real societies are linguistically diverse. Merchants develop trade pidgins. Scholars use classical languages. Nobility marks status through speech. Regions develop dialects. Fix: Design at least three registers (formal, common, intimate). Add professional jargons for important groups. Include at least one prestige/classical language.
Integration
Inbound (feeds into this skill)
| Skill | What it provides |
|---|---|
| worldbuilding | Geographic and historical context for language spread |
| multi-order-evolution | Generational timescales for language change |
| governance-systems | Political boundaries affecting language standardization |
Outbound (this skill enables)
| Skill | What this provides |
|---|---|
| conlang | Historical context for phonology choices |
| character-naming | Naming conventions following language patterns |
| dialogue | Register variation for character voice |
| memetic-depth | Linguistic markers for cultural texture |
Complementary
| Skill | Relationship |
|---|---|
| conlang | Language-evolution provides macro history; conlang provides micro phonology. Use together for deep linguistic worldbuilding |
| memetic-depth | Language-evolution tracks structural change; memetic-depth uses linguistic markers for cultural texture |
ライセンス: MIT(寛容ライセンスのため全文を引用しています) · 原本リポジトリ
詳細情報
- 作者
- jwynia
- リポジトリ
- jwynia/agent-skills
- ライセンス
- MIT
- 最終更新
- 不明
Source: https://github.com/jwynia/agent-skills / ライセンス: MIT
関連スキル
nano-banana-2
inference.sh CLIを通じてGoogle Gemini 3.1 Flash Image Preview(Nano Banana 2)で画像を生成します。テキストから画像を生成する機能、画像編集、最大14枚の複数画像入力、Google Searchグラウンディング機能に対応しています。トリガーワード:「nano banana 2」「nanobanana 2」「gemini 3.1 flash image」「gemini 3 1 flash image preview」「google image generation」
octocode-slides
洗練されたマルチファイル形式のHTMLプレゼンテーションを生成します。6段階のフロー(概要 → リサーチ → アウトライン → デザイン → 実装 → レビュー)で構成されています。各スライドは独立したHTMLファイルとなり、iframeで読み込まれます。「スライドを作成してほしい」「プレゼンテーションを作ってほしい」「HTMLスライドを生成してほしい」「デックを構築してほしい」といった依頼や、ノート・ドキュメント・コードを洗練されたプレゼンテーションに変換する際に使用できます。
gpt-image2-ppt
OpenAIのgpt-image-2を使用して、視覚的に優れたPPTスライドを生成します。Spatial Glass、Tech Blue、Editorial Monoなど10種類のキュレーション済みスタイルに対応し、ユーザーが提供したPPTXファイルを模倣するテンプレートクローンモードも搭載しています。HTMLビューアと16:9形式のPPTXファイルを出力します。プレゼンテーション、スライド、ピッチデック、投資家向けPPT、雑誌風PPTの作成依頼などで活用してください。
nano-banana
Nano Banana PRO(Gemini 3 Pro Image)およびNano Banana(Gemini 2.5 Flash Image)を使用したAI画像生成機能です。以下の場合に活用できます:(1)テキストプロンプトからの画像生成、(2)既存画像の編集、(3)インフォグラフィックス、ロゴ、商品写真、ステッカーなどのプロフェッショナルなビジュアルアセット制作、(4)複数画像での人物キャラクターの一貫性保持、(5)正確なテキスト描画を含む画像生成、(6)AI生成ビジュアルが必要なあらゆるタスク。「画像を生成」「画像を作成」「写真を作る」「ロゴをデザイン」「インフォグラフィックスを作成」「AI画像」「nano banana」またはその他の画像生成リクエストをトリガーとして機能します。
oiloil-ui-ux-guide
モダンでクリーンなUI/UXガイダンス・レビュースキルです。新機能や既存システム(Webアプリ)に対して、実行可能なUI/UX改善提案、デザイン原則、デザインレビューチェックリストが必要な場合に活用できます。CRAP(コントラスト・反復・配置・近接)をベースに、タスクファーストなUX、情報設計、フィードバック・システムステータス、一貫性、affordances、エラー防止・復旧、認知負荷を重視します。モダンミニマルスタイル(クリーン・余白・タイポグラフィ主導)を強制し、不要なテキストを削減、アイコンとしての絵文字を禁止し、統一されたアイコンセットから直感的で洗練されたアイコンを推奨します。
axiom-hig-ref
Apple Human Interface Guidelines リファレンス — 色(セマンティックカラー、カスタムカラー、パターン)、背景(マテリアル階層、ダイナミック背景)、タイポグラフィ(標準スタイル、カスタムフォント、Dynamic Type)、SF Symbols(レンダリングモード、色、多言語対応)、ダークモード、アクセシビリティ、プラットフォーム固有の考慮事項を網羅したガイドラインです。