WeOS Documentation
WeOS is an open source Go application for building a digital twin of yourself or your business — a knowledge graph of the information from the apps and devices you use, exposed to any LLM so it can answer with your real context. By default WeOS is MCP-first: it exposes an MCP server that any MCP-compatible LLM (Claude, GPT, Gemini, Ollama) connects to. Optional built-in agent integrations (e.g. Google ADK/Gemini) are available when configured.
The Go binary does three things:
- Stores your data as a knowledge graph — resources are represented as JSON-LD entities, and relationships between them are modeled as RDF triples using ontologies like Schema.org and FOAF, so people, events, products, places, messages and the relationships between them are first-class.
- Runs the MCP server — any MCP-compatible LLM connects and queries your graph for grounded, context-rich responses.
- Optionally renders sites and APIs — the same graph can drive a static-first HTML site or a REST API when you want to publish or integrate.
Where to Start
Setting up your digital twin?
Start with the Tutorials — they walk you through running WeOS, connecting an LLM, creating content types, and customizing your site step by step.
Contributing to WeOS?
Start with the Explanation section to understand the architecture — RDF ontologies, event sourcing, projections, and how MCP ties it all together. Then use the Reference for precise API and CLI details.
Looking for a specific task?
The How-to Guides are goal-oriented recipes: install, configure, deploy, and manage resources.
Quick Links
| Section | What you’ll find |
|---|---|
| Tutorials | Step-by-step learning paths |
| How-to Guides | Task-oriented recipes |
| Explanation | Conceptual deep dives |
| Reference | CLI, API, MCP, and config specs |
| Architecture | Clean Architecture overview |
| Decision Records | Architectural decisions and rationale |