RelaMind

RelaMind is an AI-powered personal growth assistant built with Java and Spring Boot. It provides smart routing for conversations, a growth diary with semantic storage, and Manus, an advanced task assistant that orchestrates multiple tools to gather information, process data, and generate outputs such as documents or PDFs. This MCP server exposes endpoints for initiating intelligent conversations, saving and querying diary entries, and triggering Manus-driven multi-step workflows. You can leverage RAG-style historical queries, tool calling (web search, file operations, PDF generation, etc.), and sensitive word filtering to ensure safe interactions. To use it, deploy the Docker image (as described in the installation guide), then call the backend API endpoints to start a chat, create diary entries, or request Manus to plan and execute complex tasks. The system is designed to stream responses where supported, and to present tool operation details while hiding lower-level implementation details.

Prerequisites:\n- Docker and Docker Compose installed on your machine.\n- Access to a PostgreSQL instance with PgVector support and optional Cassandra for memory storage.\n- A valid API key for the AI model provider you plan to use.\n\nInstall steps (Docker deployment, recommended):\n1) Clone or download the RelaMind repository.\n2) Prepare environment variables by creating a .env file or using the provided example.\n3) Start the stack with Docker Compose (as described below).\n\nDocker-based deployment (recommended):\nbash\ngit clone https://github.com/your-username/RelaMind.git\ncd RelaMind\n# Create and edit environment variables in a .env file or set them directly in your shell\n\n# Start all services using the pre-built image (from README)\ndocker-compose up -d\n\n# Verify services\ndocker-compose ps\ndocker-compose logs -f backend\n\n\nIf you prefer building locally, you can use the docker-compose setup with --build to create images from a local Dockerfile. Ensure you configure the required environment variables (DASHSCOPE_API_KEY, GITHUB_USER, GITHUB_REPO, POSTGRES_PASSWORD, CASSANDRA_PASSWORD) before starting.\n

Tips and common considerations:\n- Ensure your AI model API key is valid and accessible from the container.\n- PgVector must be available in PostgreSQL for RAG functionality; verify your database configuration and connection strings.\n- If using Cassandra for memory, make sure the cluster is reachable and credentials are correct.\n- For production, consider securing environment variables and rotating credentials regularly.\n- The README describes two deployment modes: a pre-built image workflow (simplest) and a local build workflow. Choose the one that matches your development or production needs.\n- If you encounter streaming issues, verify that the SSE setup in the frontend and backend is properly configured and that network policies allow server-sent events.\n