Show HN: Kaya Suites – An AI Native Knowledge Base (github.com)

Kaya Suites, a new AI-native knowledge base, has been announced, designed to maintain and manage documents as a dynamic source of truth. This project features a Rust backend using the Axum framework and a Next.js 16 frontend, with components integrated through a generated TypeScript OpenAPI client. Its architecture includes a unique three-agent pipeline consisting of an orchestrator for classifying chat turns, a read-only researcher for gathering evidence, and a write-only editor for proposing edits. Notably, it enforces strict read/write isolation through compile-time tests, ensuring data integrity and security. This development is significant for the AI/ML community as it introduces a sophisticated system for collaborative knowledge management that prioritizes user approval before any changes are made, thus minimizing the risk of errors. Kaya Suites also supports pluggable storage and LLM providers, allowing compatibility with popular databases like SQLite, Postgres, and MySQL, and integrating with various language models, including those from OpenAI and Anthropic. Furthermore, its Notion-style block editor aims to provide an intuitive user experience, making document manipulation easier. Overall, Kaya Suites represents a robust, modular framework for future AI-driven applications in knowledge management.