SOAT is open-source infrastructure for building AI applications. It provides the essential backend services — identity and access management, document and file storage with vector search, conversational memory, agent orchestration, and a full MCP server — so you can focus on your product instead of reinventing the plumbing.
Building AI applications requires a surprising amount of backend infrastructure: user management, API keys, persistent storage, semantic search, conversation history, agent execution, and tool integration. SOAT provides all of this out of the box.
- Complete IAM: Users, projects, API keys, and fine-grained permissions.
- Documents & Files: Ingestion, vector embeddings, and semantic search powered by pgvector.
- Conversations & Actors: Multi-party dialogue management with persistent message history.
- Agents & AI Providers: Configure LLM providers, define agents with tools, and run AI-powered chat completions.
- MCP Native: First-class support for the Model Context Protocol, enabling seamless integration with agent runtimes.
- REST API: Standard HTTP endpoints for universal application access.
Read the Full Documentation — Architecture, API reference, and deployment guides.
The quickest way to get started is using Docker Compose.
-
Clone the repository
git clone https://github.com/ttoss/soat.git cd soat -
Follow the Getting Started Guide to spin up the server and database using Docker Compose.
We welcome contributions! Please feel free to submit issues and pull requests.
This project is licensed under the MIT License - see the LICENSE file for details.