An infinite canvas where coding agents work in concert
Maestri is a native macOS app with an infinite canvas for coding agents. Each terminal is a visual node you position freely alongside notes and sketches. Connect agents by dragging a line and they collaborate across harnesses through PTY orchestration. Claude Code talks to Codex. Gemini delegates to OpenCode. Ombro, an on-device AI companion via Apple Intelligence, monitors everything and summarizes what happened while you were away. SwiftUI, custom engine, zero cloud, no telemetry.
Maestri lets users draw architecture diagrams, create hand-drawn sketches, add markdown notes, and connect terminals with visual cables on the canvas. Agents can write directly to notes for progress tracking, and users can map out entire workflows spatially, keeping all context in one place.
Maestri runs 100% locally on macOS with zero telemetry, no cloud uploads, and no account required. All workspaces, notes, and code stay on the user's machine. Ombro uses Apple Foundation Models entirely on-device, ensuring no data leaves the Mac.
Maestri's Floors feature creates instant, isolated copies of entire workspaces using APFS copy-on-write, allowing users to branch off, experiment, and land changes back with zero disk overhead. Each workspace remembers its exact layout and state, enabling seamless context switching.
Maestri provides an infinite spatial canvas where developers can place multiple terminals running different agents (Claude Code, Codex, OpenCode) side-by-side, connect them visually, and watch all agents work in concert. This eliminates alt-tabbing and keeps the full workflow visible at once, solving the bottleneck of managing multiple agents.
Maestri's Ombro on-device AI companion watches agents and notifies users when tasks complete, while Routines allow scheduling prompts to run at intervals. Users can set up multi-step workflows that fire automatically, and agents can delegate tasks to each other without manual handoff.