Synori.AI documentation
How Synori.AI works — the workspace, the sandbox your agents run in, the models they use and the tools they reach.
Synori.AI is a macOS workspace where AI coding agents work on your projects inside isolated micro-VMs, and where you approve every risky step. These pages describe what the app does and how its parts fit together.
Start with What is Synori.AI for the big picture, or pick a topic below. Just installed the app? How to start walks through the setup before the first room.
Get started
- What is Synori.AI
A desktop workspace for AI coding agents — sandboxed execution, many models side by side and a human in control of every risky step.
- Install and update
System requirements, the macFUSE dependency, the one-line installer, what it puts on your Mac, updates and removal.
- How to start
The one-time setup before the first room — an environment, the VM tools archives with the coding agents and toolchains, and unpacking them.
Working with agents
- Activity rooms
How an activity works — participants and roles, addressing models by name, background actors, approvals, verification and rollback.
- Models and providers
Connect frontier APIs, local models and the coding CLIs you already use — and tune how each model behaves in a room.
- Tools and integrations
What agents can reach from a room — the session, a built-in browser, platform tools, GitHub, MCP, skills, hooks and messengers.
- The workspace
Documents, plans, epics, issues and artifacts — what a project knows and what it still has to do, shared by you and your agents.
Security
- Security and data flow
How Synori.AI isolates agents, controls their network access, stores secrets and decides which data leaves your Mac.