blackzhanzhan

blackzhanzhan / novel_agent

Public

Git-native AI novel writing workbench with Dify agents

32
6
80% credibility
Found May 31, 2026 at 32 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A collaborative AI writing workbench for web novel authors that imports existing novels, creates intelligent summaries, and uses specialized AI agents to help generate, review, and manage story continuations while maintaining version control.

How It Works

1
📚 Import your novel

You search for your favorite web novel by pasting a link or book name, then import it with one click to start your writing project.

2
✨ Watch the magic happen

The system reads through all your chapters and creates a smart summary that tracks characters, plot twists, and story promises automatically.

3
🤖 Set up your AI writing team

You click a button to create AI assistants that learn your novel's world, writing style, and story outline so they can help you write new chapters.

4
✍️ Generate new chapters

You ask your AI assistant to write the next few chapters based on your outline, and watch as it creates draft content that matches your story's style.

5
🔍 Review and polish

Another AI assistant checks your draft for consistency, style issues, and plot problems, then suggests fixes or you can edit manually.

6
Choose your path
✅
Approve and save

You click to merge the approved chapters into your main story, keeping everything safely stored with version history

🔄
Request changes

You send the draft back to the AI with notes, and it rewrites the problematic sections before trying again

🎉 Your story grows

With each approved chapter, your novel expands while the system remembers every version, so you can always go back if needed.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 32 to 32 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is novel_agent?

novel_agent is a Git-native AI writing workbench designed for long-form fiction authors. It treats each novel as a Git repository, enabling version control and branching for draft exploration. The system orchestrates multiple AI agents through Dify workflows--a world model agent handles worldbuilding, a continuation agent drafts chapters, a review agent checks consistency, and a style agent maintains voice. The Python backend serves a React frontend that visualizes Git branches and manages chapter workflows. Authors import books, interact with specialized agents for different writing tasks, and maintain consistency across thousands of chapters through structured AI collaboration.

Why is it gaining traction?

The Git-native approach solves a real problem: managing drafts and revisions across long novels. Rather than relying on a single AI to generate everything, the multi-agent architecture distributes responsibilities--worldbuilding, drafting, reviewing, and style consistency each get dedicated agents. This separation of concerns prevents the continuity drift common with monolithic AI writing tools. The Dify integration makes the workflow customizable and transparent, letting authors understand and adjust how agents collaborate.

Who should use this?

Fiction authors writing serialized or long-form work benefit most. Web novel writers tracking hundreds of chapters, collaborative writing teams needing version control, and authors wanting AI assistance that respects established world rules will find this most useful. It's less suited for short fiction or single-draft work.

Verdict

novel_agent represents an interesting architectural choice--Git for version control, Dify for orchestration, and multiple specialized agents for different writing concerns. The 0.800000011920929% credibility score reflects a small but active project with 32 stars. Documentation quality and test coverage remain unclear from the repository, so evaluation requires hands-on testing. For authors struggling with AI consistency across long projects, this approach warrants exploration, though maturity concerns suggest starting with non-critical work.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.