This project is an AI-powered fact-finder application that allows users to ask questions about Inkeep and get answers from its documentation. The application then automatically saves the conversation to a Google Sheet. It uses a Next.js frontend with CopilotKit for the chat interface and a Mastra server on the backend to power the AI agent.
https://www.loom.com/share/a4389c10cb4743c38f338aa75e2fe86e
- Chat Interface: A simple and intuitive chat interface to interact with the AI assistant.
- Ask Questions about Inkeep: Users can ask any question about Inkeep, and the assistant will provide a relevant answer.
- Search Inkeep Documentation: The assistant can search the Inkeep documentation to find answers to user queries.
- Save Conversation to Google Sheets: The entire conversation is automatically saved to a Google Sheet for future reference.
- Extensible Agent: The backend agent can be extended with more tools and capabilities.
- Frontend:
- Backend:
The project is a monorepo with the following structure:
next-app/: The Next.js frontend application.mastra-server/: The Mastra backend server.
To get the project up and running on your local machine, follow these steps:
- Node.js (v20.9.0 or later)
- npm
-
Clone the repository:
git clone https://github.com/your-username/your-repository.git cd your-repository -
Install dependencies:
npm install
-
Set up environment variables:
Create a
.envfile in themastra-serverdirectory and add your OpenAI API key:OPENAI_API_KEY=your-openai-api-key
-
Start the backend server:
npm run start
-
Start the frontend development server:
npm run dev
Open your browser and navigate to http://localhost:3000 to see the application in action.
- Open the application in your browser.
- Type a question about Inkeep in the chat input field.
- The assistant will provide an answer based on the Inkeep documentation.
- The conversation will be saved to a Google Sheet.