RepoLens 🔍

RAG-powered codebase exploration tool leveraging Gemini AI and Pinecone vector search.

RepoLens ingests public GitHub repositories, parses and chunks source files, generates embeddings, and allows users to perform context-aware semantic searches across complex codebases with precise file-path attributions.

---

🌟 Key Features

• ⚡ Automated Repository Ingestion: Fetches and processes public GitHub repositories directly from a URL.
• 🎯 Context-Aware Code QA: Uses Retrieval-Augmented Generation (RAG) to ground AI responses in actual repository files, eliminating hallucinations.
• 📍 Source Attribution: Every generated answer cites the specific files used as context.
• 🛠️ Fully Typed Architecture: Built with TypeScript end-to-end with strict type checks.
• 📱 Responsive Dual-Panel UI: Clean split-view layout for desktop and stacked interface for mobile devices using Tailwind CSS and shadcn aesthetics.

---

🏗️ Architecture Overview

          +-----------------------+
          |  GitHub Repository    |
          +-----------+-----------+
                      |
                      v
          +-----------------------+
          |  Ingestion Pipeline   |
          |  (File Parser/Chunker)|
          +-----------+-----------+
                      |
                      v
          +-----------------------+
          |  Vector Storage       |
          |  (Pinecone DB)        |
          +-----------+-----------+
                      |
                      v
            +-------------------+     
            | User Prompt / QA  +     
            +-------------------+
                      |
                      v
          +---+-------------------+
          | RAG Retrieval         |
          | (Gemini Embeddings)   |
          +-----------+-----------+
                      |
                      v
          +-----------+-----------+
          | Contextual Answer     |
          | & File Attributions   |
          +-----------------------+

---

🛠️ Tech Stack

• Framework: Next.js (App Router, Server Actions, API Routes)
• Language: TypeScript
• AI & ML: Google Gemini AI API (@google/genai)
• Vector Database: Pinecone (@pinecone-database/pinecone)
• Styling: Tailwind CSS, Lucide React Icons
• Deployment: Vercel

---

🚀 Getting Started

Prerequisites

Ensure you have the following installed and set up:

• Node.js: v18.x or higher
• npm / pnpm / yarn
• Pinecone Account: An active index configured for vector embeddings.
• Google Gemini API Key: Access to Gemini API.

---

Environment Variables

Create a .env.local file in the root directory and add the following keys:

# Google Gemini API
GEMINI_API_KEY=your_gemini_api_key_here
# Pinecone Vector Database
PINECONE_API_KEY=your_pinecone_api_key_here
PINECONE_INDEX=your_pinecone_index_name

Installation & Local Setup

Clone the repository

    git clone https://github.com/vaidikdubey/RepoLens
    cd repolens

Install dependencies

    npm install

Run the app

    npm run dev

Open in browser

Navigate to http://localhost:3000 to access the application.

📂 Project Structure

repolens/
├── app/
│   ├── api/
│   │   ├── chat/route.ts      # Query vector store & stream Gemini responses
│   │   └── ingest/route.ts    # Parse repo, chunk code, store embeddings
│   ├── chat/page.tsx          # Dual-panel QA Terminal interface
│   ├── layout.tsx             # Root layout & global metadata
│   └── page.tsx               # Client landing page
├── lib/                       # Gemini & Pinecone client configurations
└── public/                    # Static assets & icons

---

Screenshots

• Landing Page

• Chat Interface

---

Author

Vaidik Dubey

• GitHub: https://github.com/vaidikdubey
• LinkedIn: https://linkedin.com/in/vaidikdubey

🤝 Contributing

Contributions, issues, and feature requests are welcome! Feel free to check the issues page or submit a pull request.

---

⭐ If you found this project useful, consider giving it a star!