Connapse is an open-source, container-isolated knowledge backend designed to give AI agents persistent, searchable memory across sessions. It lets users upload, organize, and search documents (via hybrid vector + keyword search) and integrates directly with AI agents through a Model Context Protocol (MCP) server, REST API, CLI, or web interface. Connapse is self-hosted, privacy-focused, and well-suited for teams or individuals who need agents to retain and query accumulated knowledge from varied sources (like S3, Azure Blob, local drive).
Visit Connapse's official website for product details and getting started.
Comprehensive guides and API references for using Connapse.
Engage with other users and developers to share insights and troubleshoot issues.
Explore practical examples and use cases to better understand how to implement Connapse.