Lobs Mission Control

macOS command center for the Lobs multi-agent system. Manage tasks, chat with AI agents, browse memory & knowledge, monitor work, and track system health — all in one unified SwiftUI app.

Features

Core

• Command Center — Dashboard with "while you were away" summary, quick stats, and recent activity
• Chat — Real-time messaging with Lobs via WebSocket
• Command Palette (⌘K) — Fuzzy finder for tasks, projects, documents, and navigation
• Quick Capture (⌘⇧Space) — Global hotkey for instant task/memory capture

Task Management

• Tasks & Kanban — Project-based boards with drag-and-drop, filtering, and agent assignment
• Inbox — Action items and proposals from AI agents requiring human decision
• Projects — Organize work by project with active/archive states

Knowledge & Memory

• Memory — Browse, search, edit, and create personal memories (your second brain)
• Knowledge/Topics — Research workspaces with documents, notes, sources, and findings
• Documents — Browse and read reports and research deliverables

Team & Monitoring

• Team View — Real-time agent status grid showing what each agent is working on
• Work Tracker — Timeline of completed work sessions with cost tracking
• Calendar — View upcoming events and schedule overview
• Status Dashboard — System health, activity feed, AI usage & cost analytics

Tech Stack

• Framework: SwiftUI (macOS 14.0+)
• Build: Swift Package Manager
• Architecture: MVVM (AppViewModel as central state holder)
• API: REST + WebSocket to lobs-server
• Network: Connects via Tailscale for secure remote access
• Storage: Local cache + server-side persistence

Building

Prerequisites

• macOS 14.0 or later
• Xcode 15+ (for Swift compiler)
• Running instance of lobs-server

Build & Run

git clone git@github.com:RafeSymonds/lobs-mission-control.git
cd lobs-mission-control
swift build
swift run

Or use the build script:

./bin/build

Configuration

On first launch, the app guides you through onboarding:

1. Server URL — Your lobs-server address (e.g., http://100.x.x.x:8000 for Tailscale)
2. API Token — Generate on server: cd ~/lobs-server && python bin/generate_token.py mission-control

Settings are stored in ~/Library/Application Support/LobsMissionControl/config.json.

You can also configure via Settings (⌘,) after initial setup.

Architecture

┌─────────────────────────────────────────────────┐
│          Lobs Mission Control (macOS)           │
│                                                 │
│  ┌──────────────────────────────────────────┐  │
│  │         AppViewModel (State)             │  │
│  └──────────────────────────────────────────┘  │
│                      │                          │
│  ┌──────────────────────────────────────────┐  │
│  │         APIService (Networking)          │  │
│  │  • REST API calls                        │  │
│  │  • WebSocket (chat, live updates)        │  │
│  │  • JSONDecoder (.convertFromSnakeCase)   │  │
│  └──────────────────────────────────────────┘  │
│                      │                          │
└──────────────────────┼──────────────────────────┘
                       │
                   (Tailscale)
                       │
┌──────────────────────┼──────────────────────────┐
│                      ▼                          │
│               lobs-server (FastAPI)             │
│  • Task orchestration                           │
│  • Agent coordination                           │
│  • Memory & knowledge storage                   │
│  • Calendar integration                         │
│  • WebSocket relay                              │
└─────────────────────────────────────────────────┘

Key patterns:

• Views use @EnvironmentObject to access AppViewModel
• API calls routed through vm.apiService (NOT direct APIService instantiation)
• Models decoded with .convertFromSnakeCase — no manual CodingKeys for snake→camel
• WebSocket updates pushed to AppViewModel, which updates @Published properties
• Local cache (CacheManager) for offline resilience

Documentation

• ARCHITECTURE.md — App structure, data flow, key components
• CONTRIBUTING.md — Development guide, debugging, common pitfalls
• AGENTS.md — AI agent guidance and constraints
• docs/ — Bug fixes, feature documentation, implementation notes
  • See docs/README.md for full index

Contributing

See CONTRIBUTING.md for guidelines.

For AI agents working on this codebase, see AGENTS.md.

See Also

Lobs Ecosystem Documentation (in ~/self-improvement/docs/):

• LOBS_ECOSYSTEM.md — Cross-project architecture and feature matrix
• GETTING_STARTED.md — 20-30 min ecosystem onboarding
• TECH_STACK_REFERENCE.md — Technology choices and patterns
• Code Quality System — Handoffs, reviews, technical debt tracking

License

Private