Skip to content

README.en.md

Latest commit

 

History

History
169 lines (127 loc) · 6.8 KB

README.en.md

File metadata and controls

169 lines (127 loc) · 6.8 KB
DeepV-Ki Logo

DeepV-Ki

AI-Powered Wiki Generator for Code Repositories

Turn any Git repository into a beautiful, interactive Wiki in minutes | Intelligent Architecture Analysis | RAG Code Q&A

license python nextjs stars issues

中文 | English


📖 Introduction

DeepV-Ki is an open-source, AI-powered Wiki generator designed to solve the challenges of maintaining and reading code documentation. Simply enter a repository URL, and DeepV-Ki will automatically analyze the code structure, generate detailed documentation, draw architecture diagrams (Mermaid), and build an interactive knowledge base supporting RAG (Retrieval-Augmented Generation) Q&A.

Whether it's GitHub, GitLab, or Bitbucket, DeepV-Ki helps developers quickly understand complex codebases.

DeepV-Ki Interface

🌟 Features

Feature Description
📚 One-Click Wiki Generation Turn any code repository into a professional interactive Wiki, supporting 10+ languages (English, Chinese, Japanese, etc.).
🧠 Intelligent Code Analysis Uses AI to deeply understand code structure, design patterns, and core logic, automatically generating table of contents and navigation.
📊 Automatic Architecture Diagrams Automatically generates Mermaid flowcharts, sequence diagrams, and class diagrams, with support for interactive zooming and panning.
💬 RAG Code Q&A Built-in Ask feature for accurate Q&A based on actual code, supporting multi-turn conversations and streaming responses.
🕵️ DeepResearch Deep research mode that automatically generates research plans and conducts multi-round iterative investigations to output comprehensive conclusions.
🔌 Multi-Model Support Supports OpenAI, Google Gemini, Azure, AWS Bedrock, Ollama (Local), and other LLMs.
🛡️ Broad Repository Support Supports GitHub, GitLab (SaaS/Self-hosted), Bitbucket, Gerrit, and private repositories.

🚀 Quick Start

Prerequisites

  • Python 3.12+ (Backend)
  • Node.js 18+ (Frontend)
  • pnpm (Frontend Package Manager)
  • uv (Python Package Manager, Recommended)

1. Clone Repository

git clone https://github.com/OrionStarAI/DeepV-Ki.git
cd DeepV-Ki

2. Configure Environment

Copy the example configuration file and fill in the necessary API Keys (e.g., OpenAI or GitLab configuration):

cp .env.example .env
# Edit .env file
# Required: OPENAI_API_KEY (or other LLM Key)
# Optional: GITLAB_CLIENT_ID (if OAuth is needed)

3. One-Click Start

We provide a unified development environment startup script:

./start_dev.sh

After successful startup, visit:

🛠️ Configuration

DeepV-Ki supports flexible environment variable configuration. Main configuration items include:

Variable Name Description Default
OPENAI_API_KEY OpenAI API Key —
GOOGLE_API_KEY Google Gemini API Key —
DASHSCOPE_API_KEY Aliyun DashScope API Key —
OPENROUTER_API_KEY OpenRouter API Key —
GITLAB_URL GitLab Instance URL https://gitlab.com
GITLAB_CLIENT_ID GitLab OAuth App ID —
GITLAB_CLIENT_SECRET GitLab OAuth App Secret —
GITLAB_REDIRECT_URI OAuth Callback URL (Must match GitLab App config) http://localhost:8001/api/auth/gitlab/callback
SESSION_SECRET_KEY Session encryption key (must set in production) —
PORT Backend Service Port 8001
SERVER_BASE_URL Backend server URL used by frontend proxy http://localhost:8001
FRONTEND_URL Frontend URL for OAuth/SSO redirects http://localhost:3000
LOG_LEVEL Log level INFO

🔗 Detailed Guide: Having issues? Check the GitLab OAuth Configuration Guide.

For the full list of environment variables with descriptions, see the .env.example file.

🏗️ Architecture

The project adopts a modern separation of frontend and backend architecture:

  • Backend (api/): Based on FastAPI and Python. Responsible for core Wiki generation logic, RAG system (AdalFlow + FAISS), task queues, and Git operations.
  • Frontend (frontend/): Based on Next.js 15 and React 19. Provides modern UI, Mermaid diagram rendering, and streaming interactive experience.
graph TD
    User[User] --> Frontend[Next.js Frontend]
    Frontend --> Backend[FastAPI Backend]
    Backend --> LLM[LLM Service OpenAI/Gemini]
    Backend --> VectorDB[FAISS Vector DB]
    Backend --> Git[Git Service GitHub/GitLab]
Loading

📚 Tech Stack

  • Backend: Python 3.12, FastAPI, Uvicorn, AdalFlow, FAISS
  • Frontend: TypeScript, Next.js 15, React 19, Tailwind CSS 4, Mermaid.js
  • DevOps: Docker, uv, pnpm

🤝 Contributing

We welcome community contributions! If you have good ideas or found a Bug, please:

  1. Fork this repository
  2. Create your feature branch (git checkout -b feature/AmazingFeature)
  3. Commit your changes (git commit -m 'Add some AmazingFeature')
  4. Push to the branch (git push origin feature/AmazingFeature)
  5. Open a Pull Request

📄 License

This project is licensed under the MIT License.

Made with ❤️ by the DeepV-Ki Team