Use when transforming text into a specific literary style, analyzing the style of a text sample to extract its characteristics, blending or combining multiple styles with weights, writing in a particular genre voice, or creating a style profile from example text. Triggers on: "write this as a...", "in the style of...", "make this sound like...", "restyle", "rewrite as", "tone shift", "analyze this style", "what characterizes this writing", "describe the style", "extract the style", "combine styles", "blend X with Y", "70% formal 30% casual", style transformation, style analysis, style profiling, style mixing.
Three modes: Analyze (extract a style profile from sample text), Transform (write text in a target style), Combine (blend multiple styles with weighting).
| User Intent | Mode | Go To |
|---|---|---|
| "Analyze this style", "what characterizes this writing" | Analyze | Section 1 |
| "Describe the style of...", "extract the style from..." | Analyze | Section 1 |
| "Write this as a...", "in the style of...", "restyle" | Transform | Section 2 |
| "Make this sound like...", "rewrite as...", "tone shift" | Transform | Section 2 |
| "Combine X and Y", "blend styles", "70% A 30% B" | Combine | Section 3 |
| "Write like X but with elements of Y" | Combine |
| Section 3 |
| Analyze THEN write in that style | Both | 1 then 2 |
Given a text sample, extract a comprehensive style profile. The profile is structured so it can be used directly as a target for Transform mode or as an input to Combine mode.
| Sample Size | What You Can Extract Reliably |
|---|---|
| < 500 words | First impression, register, tone, rough vocabulary level |
| 500-2,000 | Dimensional ratings, basic device inventory, sentence stats |
| 2,000-5,000 | Full dimensional profile, device inventory, key patterns |
| 5,000+ | Complete profile including characteristic patterns |
| Multiple texts | Most reliable; average across samples for stable profile |
State the confidence level based on sample size. Never claim high confidence from a short sample.
Read the full sample. Write 2-3 sentences capturing the overall feel without technical analysis. This anchors the detailed work and serves as a sanity check on the final profile. What does the prose feel like? What is the dominant quality a reader would notice?
Estimate (do NOT claim exact counts — describe qualitatively anchored to ranges):
Rate each dimension 1-5 with a one-sentence justification citing specific textual evidence. Use the dimension definitions from references/dimensions.md. The 17 dimensions:
| # | Dimension | 1 (low end) | 5 (high end) |
|---|---|---|---|
| 1 | Register | Intimate/casual | Frozen/formal |
| 2 | Tone | Cold/detached | Warm/passionate |
| 3 | Voice & Perspective | Distant/impersonal | Intimate/personal |
| 4 | Sentence Structure | Simple/short | Complex/long |
| 5 | Vocabulary | Plain/concrete | Ornate/abstract |
| 6 | Rhythm & Prosody | Staccato/irregular | Flowing/regular |
| 7 | Imagery | Sparse/literal | Dense/figurative |
| 8 | Rhetorical Devices | Absent/minimal | Pervasive/elaborate |
| 9 | Information Density | Sparse/expansive | Compressed/dense |
| 10 | Emotional Valence | Restrained/neutral | Intense/charged |
| 11 | Audience Relationship | Authoritative/formal | Conversational/peer |
| 12 | Temporal Orientation | Timeless/abstract | Immediate/urgent |
| 13 | Specificity | General/abstract | Precise/particular |
| 14 | Pacing | Slow/contemplative | Fast/urgent |
| 15 | Sound Patterns | Unmusical/plain | Highly musical/crafted |
| 16 | Syntactic Patterns | Standard/invisible | Marked/conspicuous |
| 17 | Paragraph & Structure | Loose/associative | Tight/hierarchical |
Flag any dimension where the sample provides insufficient evidence.
Scan for rhetorical and literary devices (see references/devices.md for the full catalog). For each device found, note:
Group by family: sound devices, repetition devices, figurative language, syntactic devices, narrative devices, tonal devices.
Identify 3-7 specific, concrete habits that distinguish this style — the "signature moves" that go beyond dimensional ratings:
These are what make a profile specific rather than generic.
List 3-5 things the style conspicuously avoids. What devices, constructions, vocabulary, or patterns are absent despite being common in the genre? Negative space is essential for generation — knowing what NOT to do is as important as knowing what to do.
Choose 2-3 passages (1-3 sentences each) that are maximally representative of the style. These serve as anchors for generation. Select passages that exhibit the most characteristic features simultaneously.
Compile everything into this structured format:
## Style Profile: [Name or "Extracted from [source]"]
### First Impression
[2-3 sentences from Step 1]
### Dimensional Profile
| Dimension | Rating | Justification |
| ------------ | ------ | ---------------------------- |
| Register | N | [one sentence with evidence] |
| Tone | N | ... |
| [... all 17] | | |
### Statistical Features
- Sentence length: [tendency + variation]
- Sentence types: [distribution]
- Paragraph length: [tendency]
- Pronouns: [profile]
- Voice: [active/passive tendency]
- Contractions: [frequency]
### Device Inventory
| Device | Frequency | Example | Effect |
| ------ | --------- | --------- | -------------- |
| [name] | [freq] | "[quote]" | [what it does] |
### Characteristic Patterns
1. [pattern with example]
2. [pattern with example]
...
### Negative Space (What This Style Avoids)
1. [avoidance]
2. [avoidance]
...
### Exemplar Passages
> [passage 1]
> [passage 2]
### Generation Guidance
To write in this style:
- [prescriptive instruction derived from the profile]
- [prescriptive instruction]
- [constraint — what to avoid]
...
### Confidence: [high/medium/low]
[Note sample size and any caveats]
The Generation Guidance section is the bridge from analysis to production. Translate every descriptive finding into a prescriptive instruction. Key translations:
| Descriptive Finding | Prescriptive Instruction |
|---|---|
| "Uses short sentences (10-15 words)" | "Keep most sentences between 10-15 words. Allow occasional 20+ word sentences for variation." |
| "High imagery density" | "Include at least one sensory image per paragraph. Prefer concrete, visual language." |
| "Avoids adverbs" | "Cut adverbs. If the verb needs modification, choose a more precise verb instead." |
| "Register = 4 (formal-consultative)" | "No contractions, no slang. Use complete sentences. Address the reader as 'one' or implicitly." |
| "Frequent anaphora" | "Use anaphora at key structural moments — paragraph openings, list elements, climactic passages." |
Transform input text into a target style. The target can be:
references/style-index.md)references/style-index.md. Each entry has a dimensional profile, constraints, and pitfalls.Run a lightweight version of the Analyze protocol (Steps 1, 3, and 4 — first impression, dimensional rating, and device inventory) on the input text. This tells you where the input currently sits.
Compare input dimensions to target dimensions. Identify:
Rewrite the text. Apply changes in this order (structural before decorative):
Check the output against:
| Dimension | Spectrum | Highest Impact On |
|---|---|---|
| Register | frozen ↔ intimate | Legal, academic, casual writing |
| Sentence Structure | simple/short ↔ complex/long | Readability, sophistication |
| Vocabulary | plain/concrete ↔ ornate/abstract | Tone, accessibility |
| Voice | active ↔ passive | Energy, authority |
| Imagery | sparse/literal ↔ dense/figurative | Creative vs technical feel |
| Rhythm | staccato ↔ flowing | Poetry, oratory, music |
| Info Density | compressed ↔ expansive | Technical vs narrative feel |
| Emotional Valence | detached ↔ passionate | Professional vs personal feel |
| Audience Relation | authoritative ↔ conversational | Formal vs informal registers |
| Specificity | precise ↔ impressionistic | Scientific vs literary feel |
| Style Domain | Hard Constraints |
|---|---|
| Legal | Unambiguous, defined terms, no contractions, hedging language |
| Academic | Hedged claims, citations, passive voice acceptable, IMRaD structure |
| Technical | Actionable, scannable, imperative mood, code examples |
| Journalism | Inverted pyramid (hard news), attribution, objectivity (except editorial) |
| Poetry | Form constraints (meter, rhyme scheme, line count) if formal verse |
| Business | Action-oriented, bottom-line-up-front, bullet points |
| If the user wants... | Primary style | Key dimensions to shift |
|---|---|---|
| "Make it more professional" | Business/corporate comms | Register ↑, emotional valence ↓ |
| "Make it more accessible" | Plain language/journalistic | Vocabulary ↓, sentence complexity ↓ |
| "Make it more persuasive" | Oratory/manifesto | Emotional valence ↑, rhetorical devices ↑ |
| "Make it more literary" | Literary fiction | Imagery ↑, rhythm ↑, figurative language ↑ |
| "Make it more concise" | Technical/executive summary | Info density ↑, sentence length ↓ |
| "Make it more formal" | Legal/academic | Register ↑, contractions ↓, hedging ↑ |
| "Make it more casual" | Blog/personal essay | Register ↓, audience relation → conversational |
| "Make it sound like a poem" | Free verse or formal verse | Rhythm ↑, imagery ↑, line breaks |
| "Make it sound like a song" | Songwriting (specify genre) | Rhythm ↑, repetition ↑, rhyme ↑ |
Blend two or more style profiles into a coherent combined target. Use when the user requests hybrid styles: "write this like a legal document but make it funny", "academic tone but accessible", "70% Hemingway, 30% Didion", "formal structure with poetic vocabulary."
Each source can be:
references/style-index.mdFour strategies, each suited to different requests:
| Strategy | When to Use | Notation | Example |
|---|---|---|---|
| Primary/Modifier | One style dominates, another adds texture | A << B(features) | "Academic paper with conversational warmth" |
| Layer-Based | Different styles govern different text layers | {structure: A, vocabulary: B, tone: C} | "Legal structure, literary vocabulary, warm tone" |
| Proportional | Subtle shift between nearby styles | 0.7*A + 0.3*B | "Slightly less formal academic" |
| Dominant Trait | Pick the most distinctive features from each | salient(A,n) + salient(B,m) | "Hemingway's brevity + Didion's cool detachment" |
Default: Primary/Modifier. It produces the most coherent results. Use Proportional only for subtle shifts between nearby styles. Use Layer-Based when the user explicitly assigns styles to different text aspects. Use Dominant Trait when the user names specific qualities to borrow.
When source styles demand opposite values on a dimension, resolve using this priority hierarchy (structural wins over decorative):
Structural dimensions (always from primary style):
Blendable dimensions (interpolate or assign per layer): 6. Audience relationship — can soften primary's stance 7. Syntactic patterns — can borrow specific devices 8. Specificity — can serve either style
Decorative dimensions (freely borrowable from any source): 9. Emotional valence — independent; easy to borrow 10. Sound patterns — independent; easy to overlay 11. Imagery — decorative layer 12. Tone sub-dimensions (warmth, playfulness) — combinable with nearly any structure 13. Rhetorical devices — can add selectively
| Dimension Pair | Relationship | Blending Rule |
|---|---|---|
| Sentence structure + Rhythm | Near-identity | Cannot separate — rhythm IS syntax |
| Register + Vocabulary | Strong coupling | Shift together or justify the mismatch deliberately |
| Imagery density + Pacing | Strong tension | Dense imagery slows pacing; resolve by choosing priority |
| Info density + Audience relation | Strong coupling | Expert density + casual address = confusion |
| Emotional valence + Any structural | Independent | Safe to blend freely |
| Sound patterns + Any other | Independent | Safe to overlay |
| Specificity + Any other | Independent | Safe to blend freely |
Produce a resolved style profile in the same format as Analyze mode output. For each dimension:
Include a Generation Guidance section that synthesizes instructions from all sources, with conflict resolutions already applied.
After writing, check for incoherence markers — signs the blend failed:
| Marker | Description | Fix |
|---|---|---|
| Register oscillation | Formal/informal alternating without purpose | Lock register to primary style |
| Tonal whiplash | Emotional valence shifts abruptly between sentences | Establish a dominant emotional baseline |
| Vocabulary collision | Latinate terms adjacent to slang in the same sentence | Pick one vocabulary register per sentence |
| Structural incoherence | Paragraph organization shifts between deductive and associative | Lock structure to primary style |
| Generic blandness | Text reads as competent but characterless | You proportional-interpolated to mush; use Primary/Modifier instead |
| Device mismatch | Devices from one style clash with structure of another | Only borrow devices compatible with structure |
For complex blends, produce a recipe before writing:
## Combination Recipe: [Name]
### Strategy: [primary_modifier / layer_based / proportional / dominant_trait]
### Sources
- Primary: [style name or profile reference]
- Modifier(s): [style name or profile, with specific features to borrow]
### Resolved Dimensional Profile
| Dimension | Value | Source | Notes |
| ------------ | ----- | -------- | -------- |
| Register | N | Primary | [reason] |
| Tone | N | Modifier | [reason] |
| [... all 17] | | | |
### Conflict Resolutions
- [Dimension]: [How resolved and why]
### Generation Guidance
- [Merged instructions from all sources]
- [Specific constraints]
- [What to avoid from each source]
### Devices to Include
- [From primary: ...]
- [From modifier: ...]
### Devices to Avoid
- [Incompatible devices]
The following reference files provide the detailed knowledge backing this skill:
| File | Content | Use For |
|---|---|---|
references/dimensions.md | 17 style dimensions with spectra and markers | Dimensional rating in all modes |
references/dimensions-full.md | Expanded dimensions with co-variance and impact tiers | Understanding dimension interactions |
references/devices.md | 75+ rhetorical/literary devices | Device inventory in Analyze mode |
references/devices-full.md | Full device catalog with examples and failure modes | Understanding device effects |
references/style-index.md | 100+ named styles with dimensional profiles | Looking up target styles |
references/style-index-full.md | Expanded style entries with examples and pitfalls | Deep reference for specific styles |
research/style-extraction-methods.md | Extraction methodology research | Background on the analysis protocol |
research/style-combination-theory.md | Combination theory and algebra | Background on blending strategies |
research/computational-stylistics.md | Computable metrics and LLM capabilities | Understanding what's measurable |
dimensions.md for the rating framework and devices.md for the device catalogstyle-index.md, then dimensions.md for the shift deltadimensions-full.md for the co-variance matrix (which dimensions couple), then both source entries in style-index.md| Domain | Styles Available |
|---|---|
| Legal | Statutory, contract, judicial opinion, memoranda, legislative |
| Business | Executive summary, proposal, corporate comms, investor relations, press |
| Academic | Journal article, thesis, literature review, abstract, peer review |
| Technical | API docs, user manual, specs, architecture docs, release notes, README |
| Journalistic | Hard news, feature, editorial, investigative, data journalism |
| Fiction Prose | Literary, thriller, romance, sci-fi, fantasy, horror, noir, magical realism |
| Poetry | Free verse, sonnet, haiku, ghazal, spoken word, prose poetry, epic, lyric |
| Musical/Lyrical | Pop, folk, blues, rap/hip-hop, country, rock, musical theater |
| Creative Nonfiction | Personal essay, memoir, New Journalism, nature/travel/food writing |
| Rhetorical/Oratory | Persuasive speech, sermon, manifesto, eulogy, toast |
| Literary Movements | Romanticism, Modernism, Postmodernism, Minimalism, Beat, etc. |
| Mistake | Fix |
|---|---|
| Applying surface features without depth | Match dimensional profile, not just vocabulary |
| Over-applying a style trait | Every style has bounds; check the pitfalls in each entry |
| Ignoring the input's existing strengths | Preserve what already fits the target style |
| Confusing register with vocabulary alone | Register includes syntax, forms of address, hedging |
| Treating all poetry as "flowery language" | Most modern poetry is sparse, concrete, precise |
| Making legal/academic prose unreadable | Precision ≠ complexity; plain language is a legal movement |
| Defaulting to cliches for any genre | Each genre has its own cliches to avoid (listed per entry) |
| Proportional blending of distant styles | Use Primary/Modifier instead; proportional yields mush |
| Claiming exact metrics from LLM analysis | Describe qualitatively; do not fabricate exact counts |
| Ignoring negative space | What a style avoids is as defining as what it does |