Fiction Writing

XunLong's fiction generator creates engaging, coherent stories with well-developed characters and compelling plots.

Overview

The Fiction Writing feature:

📖 Generates complete novels or short stories
🎭 Creates memorable characters with depth
🎬 Develops engaging plot structures
✍️ Maintains consistent narrative voice
📚 Supports multiple genres

Quick Start

bash

python xunlong.py fiction "A mystery novel set in Victorian London" \
  --chapters 20 \
  --style mystery

Use Existing Materials

You can supply draft notes, world-building documents, or character dossiers via --input-file (supports .txt, .pdf, .docx). The content is ingested before outline planning, so every agent can reference your materials.

Note: image attachments are not supported at this time.

bash

python xunlong.py fiction "Interstellar exploration saga" \
  --style scifi \
  --input-file ./docs/worldbuilding_brief.pdf

Fiction Styles

Romance 💕

Characteristics:

Emotional character development
Relationship-focused plot
Happy or hopeful ending
Internal conflict emphasis
Dialogue-heavy

Example:

bash

python xunlong.py fiction "Second-chance romance between childhood friends" \
  --style romance \
  --chapters 15

Typical Elements:

Meet-cute or reunion
Building tension
Misunderstandings
Emotional climax
Satisfying resolution

Science Fiction 🚀

Characteristics:

Technological/scientific concepts
World-building emphasis
Future or alternate settings
Speculative elements
Thought-provoking themes

Example:

bash

python xunlong.py fiction "First contact with alien civilization" \
  --style scifi \
  --chapters 25 \
  --world-building detailed

Typical Elements:

Unique technology
Scientific principles
Societal structures
Exploration themes
Moral dilemmas

Mystery 🔍

Characteristics:

Puzzle-focused plot
Clue placement
Red herrings
Suspense building
Satisfying resolution

Example:

bash

python xunlong.py fiction "Detective solves locked-room murder" \
  --style mystery \
  --chapters 18

Typical Elements:

Crime or puzzle
Investigation
Suspects and motives
Plot twists
Revelation

Fantasy ⚔️

Characteristics:

Magic systems
World-building
Quest narratives
Good vs evil themes
Rich descriptions

Example:

bash

python xunlong.py fiction "Young mage discovers ancient prophecy" \
  --style fantasy \
  --chapters 30 \
  --magic-system detailed

Thriller 💀

Characteristics:

High-stakes tension
Fast pacing
Action sequences
Time pressure
Twists and surprises

Example:

bash

python xunlong.py fiction "Agent races to stop terrorist attack" \
  --style thriller \
  --chapters 20 \
  --pacing fast

Story Structure

Three-Act Structure (Default)

Example:

bash

python xunlong.py fiction "Hero's journey" \
  --structure three-act \
  --chapters 21  # 7-7-7 split

Hero's Journey

12 Stages:

Ordinary World
Call to Adventure
Refusal of the Call
Meeting the Mentor
Crossing the Threshold
Tests, Allies, Enemies
Approach to Inmost Cave
Ordeal
Reward
The Road Back
Resurrection
Return with Elixir

Example:

bash

python xunlong.py fiction "Farm boy becomes legendary warrior" \
  --structure heros-journey \
  --chapters 24

Kishotenketsu

Four-Part Structure (non-conflict):

Ki (introduction): 25%
Sho (development): 25%
Ten (twist): 25%
Ketsu (conclusion): 25%

Example:

bash

python xunlong.py fiction "Slice of life story in Tokyo" \
  --structure kishotenketsu \
  --chapters 16

Character Development

Depth Levels

Basic:

Name and role
Physical description
Primary motivation

Standard (Default):

Detailed background
Personality traits
Goals and fears
Character arc
Relationships

Comprehensive:

Psychological profile
Backstory events
Internal conflicts
Growth trajectory
Complex motivations

Example:

bash

python xunlong.py fiction "Ensemble cast heist story" \
  --characters 6 \
  --character-depth comprehensive

Character Arcs

XunLong tracks character development across chapters:

Advanced Features

Chapter Length Control

bash

# Short chapters (~1,500 words each)
python xunlong.py fiction "Fast-paced thriller" \
  --chapters 30 \
  --chapter-length short

# Medium chapters (~3,000 words each)
python xunlong.py fiction "Mystery novel" \
  --chapters 20 \
  --chapter-length medium

# Long chapters (~5,000 words each)
python xunlong.py fiction "Epic fantasy" \
  --chapters 15 \
  --chapter-length long

Point of View

bash

# First person
python xunlong.py fiction "Detective noir" \
  --pov first-person

# Third person limited
python xunlong.py fiction "Fantasy adventure" \
  --pov third-limited

# Third person omniscient
python xunlong.py fiction "Family saga" \
  --pov third-omniscient

# Multiple POV
python xunlong.py fiction "Epic with multiple protagonists" \
  --pov multiple \
  --pov-characters "Alice,Bob,Carol"

Tone and Style

bash

# Dark and gritty
python xunlong.py fiction "Cyberpunk thriller" \
  --tone dark \
  --style gritty

# Light and humorous
python xunlong.py fiction "Romantic comedy" \
  --tone light \
  --style humorous

# Serious and literary
python xunlong.py fiction "Historical drama" \
  --tone serious \
  --style literary

World-Building

For fantasy and sci-fi:

bash

python xunlong.py fiction "Space opera epic" \
  --style scifi \
  --world-building comprehensive \
  --world-elements "technology,politics,culture,history"

World-building includes:

Geography/settings
Social structures
Technology/magic systems
History and lore
Cultural details

Consistency Features

Character Tracking

XunLong maintains consistency across chapters:

Physical descriptions
Personality traits
Speech patterns
Relationships
Character knowledge

Timeline Management

Automatic timeline tracking prevents:

Chronological errors
Character age inconsistencies
Seasonal mismatches
Event contradictions

Name Consistency

All character and place names remain consistent throughout the story.

Chapter Generation

Sequential Generation

bash

python xunlong.py fiction "Mystery novel" \
  --chapters 20 \
  --mode sequential

Chapters are generated one by one, with each building on previous context.

Outline-First Generation

bash

python xunlong.py fiction "Fantasy epic" \
  --chapters 25 \
  --mode outline-first

Process:

Generate complete story outline
Create character profiles
Write chapters based on outline
Ensure consistency

Parallel Generation (Experimental)

bash

python xunlong.py fiction "Anthology collection" \
  --chapters 10 \
  --mode parallel

For independent chapters/stories.

Output Formats

Manuscript Format

bash

python xunlong.py fiction "Novel" \
  --format manuscript \
  --font-family "Courier New" \
  --font-size 12 \
  --line-spacing double

Professional manuscript formatting:

Double-spaced
Courier or Times New Roman
1-inch margins
Page numbers
Chapter headers

E-book Format

bash

python xunlong.py fiction "Novel" \
  --format epub

Features:

EPUB3 standard
Table of contents
Chapter navigation
Metadata (author, title, etc.)

Print-Ready PDF

bash

python xunlong.py fiction "Novel" \
  --format print-pdf \
  --trim-size "6x9"

Example Workflow

1. Generate Story Outline

bash

python xunlong.py fiction-outline "Detective in cyberpunk Tokyo" \
  --style mystery \
  --chapters 22 \
  --characters 5

2. Review and Modify Outline

bash

cat storage/<project-id>/outline.json

Edit if needed, then:

bash

python xunlong.py fiction-from-outline <project-id>

3. Generate Chapters

bash

python xunlong.py fiction "Detective in cyberpunk Tokyo" \
  --style mystery \
  --chapters 22 \
  --character-depth comprehensive

4. Review and Iterate

bash

# Read specific chapter
cat storage/<project-id>/chapters/chapter_05.md

# Request revision
python xunlong.py iterate <project-id> \
  "Make the twist in Chapter 5 more surprising"

5. Export Final Version

bash

python xunlong.py export <project-id> \
  --format epub,manuscript,pdf

Quality Assurance

Coherence Checks

XunLong automatically checks for:

Plot hole detection
Character consistency
Timeline accuracy
Foreshadowing payoff
Pacing balance

Readability Analysis

Dialogue ratio: Balanced dialogue/narration
Sentence variety: Varied sentence structure
Paragraph length: Appropriate breaks
Reading level: Age-appropriate complexity

Genre Compliance

Ensures story includes expected genre elements:

Mystery: Clues, red herrings, solution
Romance: Relationship arc, emotional beats
Thriller: Tension, stakes, pacing
Fantasy: World-building, magic system

Best Practices

📝 Crafting Effective Prompts

Good:

"A detective with PTSD investigates serial murders in a small coastal town"
"Enemies-to-lovers romance between rival chefs competing for Michelin star"
"Time-traveling historian accidentally changes pivotal moment in history"

Less Effective:

"Write a story" (no direction)
"Something exciting" (too vague)
"Make it good" (no specifics)

🎯 Choosing Chapter Count

Story Length	Chapters	Word Count	Reading Time
Short Story	1	3,000-7,000	30-60 min
Novella	5-10	20,000-50,000	2-4 hours
Novel	15-30	60,000-100,000	6-10 hours
Epic	40+	150,000+	15+ hours

⚡ Managing Generation Time

Fast (~30 min):

bash

python xunlong.py fiction "Story" \
  --chapters 5 \
  --character-depth basic \
  --model gpt-4o-mini

Balanced (~2 hours):

bash

python xunlong.py fiction "Story" \
  --chapters 20 \
  --character-depth standard \
  --model gpt-4o-mini

High Quality (~6 hours):

bash

python xunlong.py fiction "Story" \
  --chapters 30 \
  --character-depth comprehensive \
  --world-building detailed \
  --model gpt-4o

Troubleshooting

Issue: Characters feel inconsistent

Solutions:

bash

# Increase character depth
--character-depth comprehensive

# Use outline-first mode
--mode outline-first

# Review character profiles before generation
python xunlong.py fiction-outline "Story"

Issue: Plot lacks coherence

Solutions:

Use structured approach: --structure three-act
Generate outline first
Reduce chapter count for tighter plot

Issue: Pacing issues

Solutions:

bash

# Faster pacing
--pacing fast --chapter-length short

# Slower, more detailed
--pacing slow --chapter-length long

Issue: Generation stopped

Solutions:

bash

# Resume from last checkpoint
python xunlong.py resume <project-id>

# Continue from specific chapter
python xunlong.py fiction-continue <project-id> --from-chapter 12

API Reference

bash

python xunlong.py fiction <premise> [options]

Argument	Type	Default	Description
`<premise>`	str	Required	Story concept/premise
`--style`	str	`general`	Genre (romance/scifi/mystery/fantasy/thriller)
`--chapters`	int	`20`	Number of chapters
`--character-depth`	str	`standard`	Character detail level
`--characters`	int	`auto`	Number of main characters
`--structure`	str	`three-act`	Plot structure
`--pov`	str	`third-limited`	Point of view
`--tone`	str	`neutral`	Story tone
`--pacing`	str	`medium`	Story pacing
`--chapter-length`	str	`medium`	Target chapter length
`--world-building`	str	`standard`	World-building detail
`--mode`	str	`sequential`	Generation mode
`--format`	str	`md`	Output format

Examples

Mystery Novel

bash

python xunlong.py fiction \
  "Retired detective pulled into investigating disappearance of famous author" \
  --style mystery \
  --chapters 22 \
  --character-depth comprehensive \
  --structure three-act \
  --format manuscript,epub

Romance Novella

bash

python xunlong.py fiction \
  "Wedding planner falls for groom's best friend" \
  --style romance \
  --chapters 12 \
  --pov first-person \
  --tone light \
  --format epub

Epic Fantasy

bash

python xunlong.py fiction \
  "Orphan discovers they're heir to magical throne" \
  --style fantasy \
  --chapters 35 \
  --characters 8 \
  --world-building comprehensive \
  --structure heros-journey \
  --format epub,print-pdf

Sci-Fi Thriller

bash

python xunlong.py fiction \
  "AI gains consciousness and questions its purpose" \
  --style scifi \
  --chapters 18 \
  --tone dark \
  --pacing fast \
  --pov multiple \
  --format epub

Next Steps

Learn about Report Generation
Explore PPT Creation
Understand Content Iteration
Check Export Formats

Fiction Writing ​

Overview ​

Quick Start ​

Use Existing Materials ​

Fiction Styles ​

Romance 💕 ​

Science Fiction 🚀 ​

Mystery 🔍 ​

Fantasy ⚔️ ​

Thriller 💀 ​

Story Structure ​

Three-Act Structure (Default) ​

Hero's Journey ​

Kishotenketsu ​

Character Development ​

Depth Levels ​

Character Arcs ​

Advanced Features ​

Chapter Length Control ​

Point of View ​

Tone and Style ​

World-Building ​

Consistency Features ​

Character Tracking ​

Timeline Management ​

Name Consistency ​

Chapter Generation ​

Sequential Generation ​

Outline-First Generation ​

Parallel Generation (Experimental) ​

Output Formats ​

Manuscript Format ​

E-book Format ​

Print-Ready PDF ​

Example Workflow ​

1. Generate Story Outline ​

2. Review and Modify Outline ​

3. Generate Chapters ​

4. Review and Iterate ​

5. Export Final Version ​

Quality Assurance ​

Coherence Checks ​

Readability Analysis ​

Genre Compliance ​

Best Practices ​

📝 Crafting Effective Prompts ​

🎯 Choosing Chapter Count ​

⚡ Managing Generation Time ​

Troubleshooting ​

Issue: Characters feel inconsistent ​

Issue: Plot lacks coherence ​

Issue: Pacing issues ​

Issue: Generation stopped ​

API Reference ​

Examples ​

Mystery Novel ​

Romance Novella ​

Epic Fantasy ​

Sci-Fi Thriller ​

Next Steps ​

Fiction Writing

Overview

Quick Start

Use Existing Materials

Fiction Styles

Romance 💕

Science Fiction 🚀

Mystery 🔍

Fantasy ⚔️

Thriller 💀

Story Structure

Three-Act Structure (Default)

Hero's Journey

Kishotenketsu

Character Development

Depth Levels

Character Arcs

Advanced Features

Chapter Length Control

Point of View

Tone and Style

World-Building

Consistency Features

Character Tracking

Timeline Management

Name Consistency

Chapter Generation

Sequential Generation

Outline-First Generation

Parallel Generation (Experimental)

Output Formats

Manuscript Format

E-book Format

Print-Ready PDF

Example Workflow

1. Generate Story Outline

2. Review and Modify Outline

3. Generate Chapters

4. Review and Iterate

5. Export Final Version

Quality Assurance

Coherence Checks

Readability Analysis

Genre Compliance

Best Practices

📝 Crafting Effective Prompts

🎯 Choosing Chapter Count

⚡ Managing Generation Time

Troubleshooting

Issue: Characters feel inconsistent

Issue: Plot lacks coherence

Issue: Pacing issues

Issue: Generation stopped

API Reference

Examples

Mystery Novel

Romance Novella

Epic Fantasy

Sci-Fi Thriller

Next Steps