Second Brain is a modern, high-performance personal knowledge management system. It allows you to connect your notes, documents, and thoughts into a unified, AI-powered interface. Using Retrieval-Augmented Generation (RAG), it provides instant answers to your questions based on your own data, complete with accurate citations.

• Semantic Search: Find information across your entire knowledge base using natural language.
• Instant Retrieval: Get answers in seconds, backed by your local or remote documents.
• Persistence: Chat history is saved to MongoDB, allowing you to pick up where you left off.
• RAG Engine: Powered by LangChain and ChromaDB for efficient document ingestion and retrieval.
• Privacy First: Your data remains yours. Processes are designed for secure, local-first workflows where possible.
• Modern UI: A sleek, reactive interface built with Next.js, Framer Motion, and Tailwind CSS.

• Frontend: Next.js 15+, React 19, Tailwind CSS, Framer Motion
• AI/LLM: LangChain, Google Gemini API, Vercel AI SDK
• Database: MongoDB (Session storage), ChromaDB (Vector store)
• ORM: Prisma
• Deployment: Vercel

• Node.js 20+
• MongoDB instance (local or Atlas)
• ChromaDB instance
• Google Gemini API Key

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

MONGODB_URI=your_mongodb_uri
GOOGLE_GENERAI_API_KEY=your_gemini_api_key
CHROMA_URL=your_chromadb_url

1. Clone the repository:

   git clone https://github.com/Yuvadi29/Second-Brain.git
   cd second-brain

2. Install dependencies:

   npm install

3. Run the development server:

   npm run dev

4. Open the app: Navigate to http://localhost:3000.

1. Ingestion: Place your documents (PDF, Markdown, Text) in the knowledge/ directory.
2. Processing: The system chunks the text, generates embeddings, and stores them in ChromaDB.
3. Retrieval: When you ask a question, the system finds the most relevant chunks.
4. Generation: The LLM uses the retrieved context to generate a precise answer with citations.

Second Brain includes a GitHub Actions workflow that automatically ingests new documents whenever you push changes to the knowledge/ directory.

The convertToEmbeddings.yml workflow:

• Triggers on any push to the knowledge/ directory
• Detects all changed files automatically
• Sends files to the ingestion API endpoint
• Embeds content into ChromaDB for instant searching

1. Add your ingestion API key to GitHub Secrets:

   • Go to your repository Settings → Secrets and variables → Actions
   • Create a new secret named INGEST_API_KEY

2. Update the ingestion API URL in .github/workflows/convertToEmbeddings.yml:

   INGEST_API_URL: your_api_endpoint_url

3. Ensure your ingestion server is running and accessible

4. Push changes to the knowledge/ directory to trigger the workflow

• Check the Actions tab in your GitHub repository to see workflow execution logs
• Each run shows which files were processed and any ingestion status

For a more detailed guide on setup, architecture, and extension, visit the Documentation Page (available locally).