Revornix

Revornix is an open-source, local-first AI information workspace that helps you capture fragmented inputs, structure them into coherent knowledge, and deliver outputs through reports, visuals, podcasts, and automated notifications. The MCP components are designed to let you interact with the system programmatically: you can run the full stack via Docker Compose and then leverage the MCP API/clients to manage data, trigger transformations, and receive updates. The stack includes a FastAPI backend, a Next.js frontend, an async Celery worker, and services for embedding, graph reasoning, and notifications, all working together to provide a seamless knowledge workflow. Expect capabilities such as multi-format ingestion, advanced content transformations, vector-based retrieval, graph-context reasoning, automated podcast generation, AI-assisted illustrations, and integrated notifications for delivered outputs.

Prerequisites:\n- Docker and Docker Compose installed on your machine.\n- Git for cloning the repository.\n- Optional: you may want a local environment for development, but Revornix is designed to run via Docker Compose for quick setup.\n\nInstallation steps:\n1) Clone the repository:\nbash\ngit clone https://github.com/Qingyon-AI/Revornix.git\ncd Revornix\n\n2) Copy environment templates if you plan to customize local services (optional):\nbash\ncp .env.local.example .env.local\n 3) Start the full stack with Docker Compose (this will pull images and start all services described in docker-compose.yaml):\nbash\ndocker compose -f ./docker-compose.yaml up -d\n 4) If you plan to run locally without Docker, refer to the individual service start commands in the Quick Start section of the README, but note that a Docker-based setup is the recommended path for compatibility.\n\nAfter startup, access the UI at http://localhost:3000 and the API at the configured ports in docker-compose.yaml.\n

Tips and caveats:\n- For data isolation and consistency, Revornix recommends running via Docker Compose as described; ensure your .env.local (or equivalent) matches the services you intend to enable (Postgres, Redis, Milvus, MinIO, etc.).\n- If you already have dependencies like databases or vector stores running locally, you may need to adjust ports or disable conflicting services in docker-compose.yaml.\n- SECRET_KEY should remain consistent across services to avoid cross-service authentication failures when deploying manually.\n- For development, you can interact with individual services (api, web, celery-worker) using the commands provided in the Quick Start; for production-like runs, rely on the docker-compose setup.\n- The MCP integration is built-in, so you can use the MCP client/server flow to manage data pipelines, embed content, and trigger transformations across the stack.\n- Ensure your environment supports the required Python versions and tooling if you decide to run services outside Docker.