DeepWiki-Open automatically generates interactive wikis for GitHub, GitLab, and Bitbucket repositories using AI. It's designed for developers and teams seeking to quickly document and understand their codebase, offering features like instant documentation, private repo support, AI-powered analysis, and a chat interface for code-related queries.

How It Works

The system clones and analyzes a given repository, creating code embeddings for efficient retrieval. It leverages AI to generate documentation and Mermaid diagrams visualizing code relationships. A Retrieval Augmented Generation (RAG) approach powers an "Ask" feature, allowing users to query the repository's code, and a "DeepResearch" feature enables multi-turn investigations into complex topics.

Quick Start & Requirements

• Docker: docker-compose up after creating a .env file with GOOGLE_API_KEY and OPENAI_API_KEY.
• Manual:
  1. Create .env with API keys.
  2. Backend: pip install -r api/requirements.txt then python -m api.main.
  3. Frontend: npm install (or yarn install) then npm run dev (or yarn dev).
• Prerequisites: Google API key (Google AI Studio), OpenAI API key (OpenAI Platform).
• Links: Discord

Highlighted Details

• AI-powered code analysis and documentation generation.
• RAG-powered chat interface for querying repository code.
• "DeepResearch" feature for multi-turn, in-depth code investigations.
• Support for private repositories via personal access tokens.
• Automatic generation of Mermaid diagrams for architecture visualization.

Maintenance & Community

The project encourages contributions via issues and pull requests. Community support is available via Discord.

Licensing & Compatibility

MIT License. Permissive for commercial use and closed-source linking.

Limitations & Caveats

Requires API keys for Google and OpenAI, incurring potential costs. Performance may vary with repository size.