osaurus: what it is, what problem it solves & why it's gaining traction

What it solves

Osaurus is a local AI harness for macOS that allows users to own their AI context, memory, and tools. It solves the problem of AI data being locked in cloud servers by keeping the identity, memory, and tool execution layer on the user's machine, while remaining model-agnostic so users can swap between local and cloud models without losing their personal AI state.

How it works

Built natively in Swift for Apple Silicon, Osaurus acts as an intermediary layer between the user and various LLMs. It manages specialized agents with their own prompts and memory, and provides a secure sandbox (an isolated Linux VM via Apple's Containerization framework) where agents can execute code safely. It uses a three-layer memory system (identity, pinned facts, and session episodes) to maintain context and an on-device privacy filter to scrub sensitive data before it is sent to cloud providers.

Who it’s for

Mac users (macOS 15.5+ with Apple Silicon) who want a private, local-first AI environment capable of autonomous agent execution, secure code running, and integration with their local files and system tools.

Highlights

• Model Agnostic: Supports local MLX inference (Gemma, Llama, etc.), Liquid AI's LFMs, Apple Foundation Models, and major cloud APIs (OpenAI, Anthropic, etc.).
• Secure Sandbox: Agents run code in an isolated Alpine Linux VM with per-agent isolation.
• Privacy-First: Includes an on-device classifier to redact PII before cloud transmission and supports end-to-end encrypted communication between agents.
• MCP Integration: Functions as both a Model Context Protocol (MCP) server and client, allowing it to aggregate tools from remote providers.
• Native Performance: Written in Swift without Electron, utilizing Apple's Neural Engine for on-device transcription and optimized MLX runtimes.