Skip to main content

AI-powered narrative generation system implementing Software 3.0 methodology - The Soul of Your Story's Blueprint

Project description

Storytelling: Turn Your Ideas Into Complete Narratives

Transform a simple story idea into a fully-written, multi-chapter narrative. WillWrite is an AI-powered storytelling partner that handles the heavy lifting—outline generation, chapter planning, scene development, and revision—so you can focus on your creative vision.

What It Does

Create complete stories in minutes. You provide a story prompt. The system generates:

  • A structured outline with plot, characters, and themes
  • Individual chapters with detailed chapter outlines
  • Four polished scenes per chapter
  • Multiple revision passes for consistency and quality
  • Session persistence—pause and resume anytime

Powered by advanced AI orchestration. Built on LangGraph and LangChain with multi-LLM provider support, knowledge-base integration, and intelligent retrieval-augmented generation (RAG).

Getting Started

Installation

# Basic installation
pip install storytelling

# With full features (web fetching, local embeddings, cloud logging)
pip install storytelling[all]

Your First Story

# Create a prompt file
echo "A detective discovers a mysterious library" > my_prompt.txt

# Generate your story
storytelling --prompt my_prompt.txt --output my_story.md

Your completed story appears in my_story.md.

Using Your Own LLMs

# With local Ollama
storytelling --prompt prompt.txt \
  --initial-outline-model "ollama://mistral@localhost:11434"

# With Google Gemini
storytelling --prompt prompt.txt \
  --initial-outline-model "google://gemini-2.5-flash"

Key Features

Flexible AI Providers

  • Google (Gemini, PaLM)
  • Ollama (local models)
  • OpenRouter (community models)
  • Custom endpoints

Knowledge-Aware Generation

  • Integrate your own knowledge bases
  • Web content retrieval
  • Semantic context enhancement
  • Consistent worldbuilding

Session Management

  • Pause and resume at any stage
  • Checkpoint-based recovery
  • Narrative branching
  • Full generation history

Production Ready

  • Comprehensive logging with Langfuse
  • Type-safe with Pydantic
  • Configurable for various use cases
  • Extensible architecture

Command Line

storytelling --help                    # See all options
storytelling --prompt file.txt         # Generate story
storytelling --list-sessions           # See your past sessions
storytelling --resume <session-id>     # Continue a session

Documentation

  • Full Setup Guide: See scripts/init.sh for environment setup
  • Architecture: Check rispecs/ directory for detailed specifications
  • Specifications: RISPECS.md contains implementation architecture
  • Prompts & Models: rispecs/Prompts.md and rispecs/LLM_Provider_Specification.md

Development

# Setup development environment
./scripts/init.sh

# Run tests
make test

# Check code quality
make lint

# Build and release
make release-check

License

CC0-1.0 License - See LICENSE

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

storytelling-0.2.8.tar.gz (54.4 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

storytelling-0.2.8-py3-none-any.whl (56.1 kB view details)

Uploaded Python 3

File details

Details for the file storytelling-0.2.8.tar.gz.

File metadata

  • Download URL: storytelling-0.2.8.tar.gz
  • Upload date:
  • Size: 54.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for storytelling-0.2.8.tar.gz
Algorithm Hash digest
SHA256 05f659a46ebd14fbe07cc7b93c737ff7eb423affc15d138d8dad001bc1b530de
MD5 739c035fef8c193f2c1c0a2bbc123761
BLAKE2b-256 96548dc3f74435a36d330766ab18947923229a806b972927bd63706165572f1a

See more details on using hashes here.

File details

Details for the file storytelling-0.2.8-py3-none-any.whl.

File metadata

  • Download URL: storytelling-0.2.8-py3-none-any.whl
  • Upload date:
  • Size: 56.1 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.14

File hashes

Hashes for storytelling-0.2.8-py3-none-any.whl
Algorithm Hash digest
SHA256 acc0b66d52f2f34efa8293efae23a67f877efb3573a5417e1585d2579a2ee443
MD5 62ee070d7a022e97161a99d500b03c0f
BLAKE2b-256 c755449ecb7eda01c50447e6d0537276eb6ee61d84f0ba0020a97f750bee7422

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page