NovelToGame

Project page: https://zenstory.ai/novel-to-game · All ZenStory AI projects: https://zenstory.ai/projects

A source-grounded novel-to-game workflow: adaptation design, target-runtime builds, and evidence-based QA.

Validate Latest release License GitHub stars

NovelToGame is an open-source Agent Skills toolkit for Claude Code, Codex, and Kimi Code. It turns novel adaptation into a staged workflow: source analysis, concept selection, world and art direction, implementation, and runtime QA.

Bring a novel in any language and choose the target runtime. Generated artifacts follow the requested language; build and QA stay on the chosen platform instead of silently falling back to an easier substitute.

中文 · Play Online · Quick Start · Workflow · Skills · Artifacts · Contributing

Play Online

Three playable adaptations, each available in a browser right now and linked to the case study behind it: source provenance, concept trade-offs, game and art direction, runnable source, and evidence from the playable paths.

Journey to the West · Three Borrowings of the Banana Fan

Wukong's party and the Bull Demon King's formation remain fully visible above a separate light-silk command tray on the Jilei Mountain stage

One wave of a fan blew you fifty thousand li. Take the mountain back one turn at a time.

Command Wukong's party through the three borrowings of the Banana-Leaf Fan: read the five-element wheel, follow a fire-vein treasure map, decide when to press deeper or bank the haul, transform your way in where force will not work, and turn a demon king who outclasses you into a rainstorm over the Mountain of Flames.

Play in browser · Read the case study · design estimate: 45–90 min · all ages · playable prototype

Jin Ping Mei · Ledger of Desire

Five women of the Ximen household face the player across the household ledger

Choose whose door you enter tonight. Find out whose door knocks in the morning.

Twenty days, five courtyards. Keep silver, influence, reputation, exposure, and household strain in balance; respect each woman's terms; build trust through shared crises; and face a final ledger shaped by what everyone chose and remembers.

Play in browser · Read the case study · design estimate: 60–90 min · 18+ · playable prototype

Project Plateau · The Lost World · 3D

A real-time first-person 3D field-photography game adapted from Arthur Conan Doyle's The Lost World. Cross a connected plateau, observe a living Iguanodon family, expose four glass plates under aerial pressure, and return with the views that survived.

Play the full expedition on desktop, or watch the 15-second gameplay preview on other devices.

https://github.com/user-attachments/assets/27819247-4e4d-4bf0-8f0f-43d4125c4d45

Play in your browser — no install · Read the case study · Share feedback · 1–3 min run · desktop WebGL2 · playable prototype

Why NovelToGame

A one-line “turn this book into a game” prompt often produces a generic reskin or a clickable plot summary. NovelToGame keeps the adaptation traceable and gives each major decision a clear owner:

  • Source-grounded adaptation: extract rules, spaces, character agency, conflicts, and visual anchors with citations;
  • Real game design: turn source evidence into player verbs, systems, levels, feedback, failure, and outcomes;
  • Target-runtime delivery: build for the approved platform or engine without implementation silently redesigning the game;
  • Optional, restrained voice: synthesize only selected high-value lines at build time, keep subtitles and mute fallbacks, and never send the whole novel to a TTS provider by default;
  • Evidence-based QA: verify startup, rendering, input, the core loop, an outcome, restart, and explicit limitations in the tested runtime.

Choose the first design task

Use the quick-start guide to scope a first adaptation, then the meaningful-choice guide when choices need lasting consequences rather than different button wording.

  • Scoped first adaptation: separate authorized source facts, author-approved additions, and unresolved questions; keep the first slice small without changing the chosen engine or platform.
  • Meaningful choices: state each player option, available evidence, cost, and visible outcome, then name the later scene that reads the resulting state instead of silently resetting it.

Planning-only brief for an agent with the skills installed:

Using my authorized source, plan one small gameplay-design slice for my target engine.
Keep the choices and outcomes bounded; show each option's evidence, cost, visible effect, and where a later scene uses its state.
Label allowed additions and unresolved questions. Deliver design notes only—do not build, run QA, or claim a finished runtime.

Quick Start

1. Install the seven skills

Agent CLI Install Invoke
Claude Code npx skills add zenstory-ai/novel-to-game -g -y -a claude-code -s '*' /novel-to-game
Codex npx skills add zenstory-ai/novel-to-game -g -y -a codex -s '*' $novel-to-game
Kimi Code npx skills add zenstory-ai/novel-to-game -g -y -a kimi-code-cli -s '*' /skill:novel-to-game

Install adapters for all three CLIs on the same machine:

npx skills add zenstory-ai/novel-to-game -g -y -s '*' \
  -a claude-code -a codex -a kimi-code-cli

Cloning the repository also enables project-local skill discovery in all three CLIs.

2. Start an adaptation

Give the agent a novel file, directory, or link:

Use novel-to-game quick to adapt this novel into a fully playable game.
Recommend the target platform, genre, and engine from the source, and keep the first build to about 15 minutes.
Let the player enter the world as an original character with a new playable route through its conflict.

When you want an interactive story rather than a systems game, say so. That locks the narrative-led experience profile, so concept, design, and QA judge continuous scenes, character dialogue, testimony, and key choices instead of applying rounds, cards, and resource bars:

Use novel-to-game quick to adapt this novel into an interactive story.
Carry the experience with continuous scenes, character dialogue, testimony, and key choices.
Keep variables as hidden causal tags rather than a visible stat panel.
Key choices must change later scenes, character attitudes, and the ending, and be named back in later text.

The narrative track keeps the same agency and runtime-evidence requirements, without forcing a fixed progression arc: questioning people, pressing contradictions, and changing attitudes can carry the player loop.

quick is the low-friction option: the agent drafts sensible defaults, asks only about materially branching or safety-sensitive choices, compares meaningful alternatives, and continues through design, build, and QA. Already selected directions are validated, not reopened to fill a quota. QA covers real startup, rendering, input, a complete loop, an outcome, restart, and explicit limitations. Diagnosis and reruns are allowed; the final record must come from one complete run. No human-playtest gate or separate approval report is required. Choose director when you want to pick the concept yourself.

Claude Code

/plugin marketplace add zenstory-ai/novel-to-game
/plugin install novel-to-game@novel-to-game-skills
/novel-to-game:novel-to-game quick

Codex

codex plugin marketplace add zenstory-ai/novel-to-game
codex plugin add novel-to-game@novel-to-game-skills

Kimi Code 0.27 or newer

/plugins install https://github.com/zenstory-ai/novel-to-game
/reload
/skill:novel-to-game quick

Workflow

The orchestrator locks PRODUCT_BRIEF.md, then hands the adaptation through separately owned decisions. Concept, experience/level design, and art direction remain distinct. After world design, a risk-matched whitebox tests the hardest causal, systemic, spatial, or control question before art production; its findings return to the design owner rather than becoming a new QA gate.

Novel → Source analysis → Concept → World design → Risk-matched whitebox ↺ → Art direction → Production build → QA → Playable game

The whitebox runs only the narrow check needed for its declared risk. The production build targets the chosen runtime and prepares one authoritative verification entry point. QA may diagnose, fix, and rerun; its final record binds all six player-visible effects to the same complete execution, without reusing stale passes. Capability-specific regression checks run only when that capability is adopted. No human-playtest gate or duplicate QA report is required. Source-code provenance, public hosting, marketing, rights, subjective fun, and publication quality are not machine-proven by this QA record.

Skills

Skill Responsibility
novel-to-game Confirm requirements, choose a mode, orchestrate stage handoffs, and recover progress
novel-game-analyze Extract cited rules, verbs, spaces, agents, systems, and signature moments
game-concept Compare meaningful alternatives, reject invalid options, and select or validate a direction
game-world-design Define the player promise, core loop, world response, systems, levels, failure, and outcomes
game-art-direction Define camera, composition, visual grammar, colour, light, materials, HUD, motion, and sound
game-build Build a risk-matched whitebox, then implement the approved production candidate without redesigning it
game-qa Verify commands, states, screenshots, and real play paths without overstating subjective results

Artifacts

Each run creates a compact, self-contained adaptation workspace:

game-adaptations/<project>/
  PRODUCT_BRIEF.md
  analysis/SOURCE_BIBLE.md
  concepts/CONCEPT.md
  design/GAME_DESIGN.md
  design/ART_DIRECTION.md
  build/BUILD_BRIEF.md
  build/app/
  qa/verification.json
  _progress.md

Core design documents remain independent of any single model or game engine. The approved target runtime determines the implementation and QA environment.

Contributing

Reproducible bugs, skill gaps backed by evidence, and example proposals that demonstrate a distinct adaptation lesson are welcome. Read the contribution guide and use the repository's structured issue and pull request templates.

License

NovelToGame is released under the MIT License.

Acknowledgments

Thanks to the linux.do community for early feedback and support.

Part of ZenStory AI

This project is maintained by ZenStory AI — open-source, agent-native tools for creating, adapting and producing stories (GitHub org: zenstory-ai). Sibling projects:

Project What it does
oh-story-claudecode Web-fiction writing skill pack: chart scanning, deconstruction, drafting, de-AI-flavor, covers
drama-skills AI short-drama / motion-comic suite: scripts, assets, storyboards, image & video prompts, review
novel-to-game Agent skills for source-grounded novel adaptation, target-runtime builds, and evidence-based QA (this repo)
video-recap-skills Create Chinese-narration recaps from supported video files, with optional editable JianYing/CapCut draft export
oh-story-dsh Community DeepSeek Harness plugin with novel, short-drama, game and video-recap workbenches
zenstory Chat-to-create AI novel-writing workbench (app.zenstory.ai)