skaldlabs/skald

A production-ready RAG system handling end-to-end ingestion (document parsing, chunking, embedding, vector storage) and retrieval (query rewriting, semantic search, LLM chat with history). Provides configurable defaults for vector search parameters, reranking strategies, and model selection, with built-in evaluation tools for performance tuning. Supports Python, Node.js, Go, Ruby, PHP, C#, and MCP through dedicated SDKs, and can run fully self-hosted with zero third-party dependencies including local LLM inference and embeddings.

550 stars. Actively maintained with 1 commit in the last 30 days.