Kaig is an open-source collection of AI demo applications that use SurrealDB as their data layer. Each demo is a self-contained project you can run locally to explore a specific AI pattern.
What you will find
The repository includes demos covering:
Retrieval-augmented generation (RAG) - embedding documents into SurrealDB's vector indexes and using similarity search to ground LLM responses in real data.
AI agents - autonomous agents that read from and write to SurrealDB as part of their tool-use loop.
Knowledge graphs - building and querying graph structures that give AI models structured context.
Getting started
Clone the repository and follow the README in the demo you want to explore:
git clone https://github.com/surrealdb/kaig.git
cd kaigEach demo directory contains its own setup instructions, dependencies, and a walkthrough of the code.