index.md

Open RLM Memory Docs

Open RLM Memory is a local-first memory server for AI agents.

• Backend and frontend are served by one FastAPI app on port 8000.
• Memory data is stored in PostgreSQL with pgvector.
• Search caching uses an app-owned PostgreSQL table: pg_cache.
• Identity and data isolation use X-Memory-Namespace (no OAuth/JWT/Auth0).
• LLM and embedding calls use an OpenAI-compatible endpoint (LM Studio by default).

Sections

• Getting Started
  • Installation
  • Configuration
  • Local Development
• Guides
  • MCP Inspector Testing
  • Run Tests
  • Setting Up Monitoring
  • Deploying to Production
• Concepts
  • Architectural Overview
  • Security Model
• Reference
  • API Endpoints
  • Data Model
  • Environment Variables
  • Commands & Scripts
  • Dependencies
  • Docker Reference