lingfengQAQ/webnovel-writer

Project overview

It addresses a specific failure mode of AI novel generation—forgotten settings, dropped plot threads, and broken consistency at scale—by committing chapter facts into queryable state, indices, summaries, long-term memory, and vectors.

Project type

AI Agent · RAG · Prompt Engineering

Use cases

Documents & Office · Knowledge Q&A

Deployment

Refer to project documentation

License

GPL-3.0

Best for

• Fiction creators writing long-form serialized novels who need settings, plot threads, and rules to remain consistent across hundreds of chapters and millions of words, and who are willing to work through a structured contract-and-commit workflow.
• Users already in the Claude Code environment who can run skill commands and configure external embedding/reranking APIs for semantic recall.

Key capabilities

• Defines contracts and commits for chapters, serving as the single source of truth for facts before writing and after committing.
• Prepares context, drafts text, runs multi-dimensional reviews, polishes, extracts facts, and drives state/index/memory projections for a completed chapter.
• Automatically synchronizes newly committed facts into queryable state, indices, chapter summaries, long-term memory, and vectors.
• Reviews written chapters across consistency, pacing, out-of-character detection, and engagement metrics.
• A pre-packaged dashboard to browse project status, entity graphs, chapter contents, and engagement data without running npm build.
• Manages vector indices and retrieval status, falling back to BM25 keyword search if embedding API keys are not provided.
• Includes 37 built-in Chinese web novel genre templates, with support for combining multiple genres.
• Checks directories, files, databases, RAG, dependencies, and dashboard artifacts to provide stage-aware status and repair suggestions.

Limitations and risks

• The project relies on the Claude Code environment to run skill commands and execute its automated pipeline, so it cannot operate as a standalone tool outside that environment.
• Achieving optimal semantic recall requires external paid API keys for embeddings and reranking; without them the system falls back to BM25 keyword search. The documented cost dependency is paid_service_required, and the data boundary is external optional—so a fully local, offline data flow is not established.
• GPU requirements and telemetry practices are not documented, and minimum hardware requirements are unknown; verify these against your own setup before relying on the pipeline.

Getting started

• Setup is rated medium because it requires the Claude Code environment, installing Python dependencies, and configuring external RAG APIs for optimal semantic recall. The documented first-success path is: install the plugin via the Claude Code Marketplace, install Python dependencies, initialize a book via /webnovel-init, and configure RAG by copying .env.example to .env.

Evidence and sources

• GitHub project description
• README
• README
• README
• README

Source updated at: 2026-07-14T18:05:18Z

Open original GitHub project