This project provides a chatbot for querying a Notion knowledge base, leveraging OpenAI, LangChain, and Pinecone. It's designed for users who want to build a question-answering system over their private documentation without extensive coding.

How It Works

The system ingests Notion documents by exporting them as Markdown and CSV. These documents are then converted into numerical embeddings using OpenAI's models. These embeddings are stored in a Pinecone vector database, enabling efficient similarity searches to find relevant information for answering user queries.

Quick Start & Requirements

• Install dependencies: pnpm install
• Set up environment variables: Copy .env.example to .env and add OPENAI_API_KEY, PINECONE_API_KEY, and PINECONE_ENVIRONMENT.
• Configure Pinecone index name in config/pinecone-index.ts.
• Ingest data: npm run ingest
• Run the application: npm run dev
• Requires Node.js, pnpm, OpenAI API key, and Pinecone account.

Highlighted Details

• Builds a chatbot for Notion knowledge bases.
• Utilizes LangChain for orchestration.
• Employs Pinecone for vector storage and similarity search.
• Supports data ingestion from Notion exports (Markdown/CSV).

Maintenance & Community

Inspired by notion-qa. No specific community links or contributor details are provided in the README.

Licensing & Compatibility

The README does not specify a license. Compatibility for commercial use or closed-source linking is not detailed.

Limitations & Caveats

The project is presented as a personal project without explicit mention of testing, production readiness, or ongoing maintenance. The lack of a specified license may pose compatibility issues for commercial applications.