📌 Features
Agentic AI Code Search – Search codebases using natural language queries.
Context-Aware Navigation – Understands repository structure & dependencies.
Full Stack Code Analysis – Works with frontend, backend, and database layers.
AI-Powered Documentation – Auto-generates explanations for code snippets & modules.
Feature Discovery Mode – Find where a specific functionality is implemented.
VS Code Integration – Native UI experience inside VS Code editor.
Multiple Code Repositories Support – Switch between projects effortlessly.
🛠️ Tech Stack Layer Technologies Used Frontend React, TailwindCSS, TypeScript Backend Node.js, Express.js AI Layer OpenAI GPT-4o / GPT-5 API, LangChain Database PostgreSQL (code metadata), Pinecone/Weaviate (vector DB) Extension API VS Code Extension API Deployment Vercel (frontend), Railway/Render (backend) 📂 Folder Structure . ├── extension/ # VS Code extension source code ├── backend/ # Node.js + Express API ├── frontend/ # React + Tailwind UI (Webview) ├── scripts/ # Data ingestion & indexing scripts ├── docs/ # Documentation & architecture diagrams └── README.md
⚙️ Installation & Setup 1️⃣ Clone the Repository git clone https://github.com/your-username/agentic-ai-vscode-extension.git cd agentic-ai-vscode-extension
2️⃣ Install Dependencies cd backend && npm install cd ../frontend && npm install cd ../extension && npm install
3️⃣ Configure Environment Variables
Create a .env file in backend/ with:
OPENAI_API_KEY=your_openai_key DATABASE_URL=postgres_connection_string VECTOR_DB_URL=pinecone/weaviate_endpoint
4️⃣ Run the Development Servers
cd backend && npm run dev
cd ../frontend && npm run dev
cd ../extension && npm run watch
💡 Usage
Open VS Code & load your project.
Open the AI Code Search panel from the sidebar.
Enter a natural language query (e.g., "Where is user authentication handled?").
View results with highlighted matches & AI-generated explanations.
Click on any result to jump directly to the file and line number.
📊 Architecture Diagram [ VS Code Extension ] | v [ React Webview UI ] <--> [ Node.js Backend API ] | v [ Vector DB (Pinecone) + PostgreSQL ] <--> [ OpenAI GPT + LangChain ]
🛠️ Future Improvements
Offline AI Mode using local LLMs.
Refactoring Suggestions for improving legacy code.
Integration with GitHub/GitLab APIs for live repo analysis.
Advanced Semantic Search using hybrid BM25 + embeddings.
🤝 Contributing
Pull requests are welcome! Please open an issue first to discuss major changes.
📜 License
MIT License – feel free to use and modify with attribution.