Stars
Forks
Watchers
Developer links
AMUX
To coordinate swarms of autonomous coding agents across complex software projects, AMUX provides a concurrency-safe kanban workspace with real-time thought stream inspection and live steering controls. Engineers assign development tasks through an interactive board where autonomous worker processes claim cards using compare-and-swap mutexes to eliminate race conditions between parallel sessions. Operators monitor active reasoning streams, inspect streaming terminal outputs, and inject guidance commands mid-turn without interrupting background execution. A cryptographic origin router relays messages between agents, enabling workers to delegate sub-tasks, request automated peer code reviews, and share environment state safely. Built-in schedulers trigger recurring automated maintenance routines, dependency updates, and bug triage sweeps on cron-like timers or autonomous loop cadences. Teams can switch language models dynamically on active sessions to tackle complex architecture challenges with high-capability reasoning engines while conserving resources on routine tasks. An integrated self-healing supervisor watches over agent runtimes, automatically compacting context windows, restarting crashed terminal sessions, and resuming pending instructions seamlessly. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. MIT licensed.
Benefits
- Collision-Free Parallel Task Execution
- Eliminate duplicate effort across large autonomous agent fleets through an atomic kanban task board that uses compare-and-swap claiming mechanisms to prevent multiple workers from touching the same issue.
- Interactive Mid-Flight Agent Steering
- Redirect active coding workflows, inject terminal keystrokes, and answer clarification prompts directly from responsive desktop or smartphone interfaces without aborting running processes or corrupting session history.
- Verified Inter-Agent Communication
- Enable collaborating agents to coordinate complex engineering tasks, share scoped working memory, and conduct peer code audits using origin-stamped messaging channels managed securely by the orchestrator.
- Automated Self-Healing Fleet Reliability
- Ensure continuous unattended operation with an active watchdog service that automatically compacts token context, recovers crashed terminal sessions, and safely retries uncompleted queue items after transient failures.
Features
- Atomic Kanban Board
- Dispatches issues through compare-and-swap claiming rules that guarantee unique task ownership across parallel agent workers.
- Multi-Agent Tiled Workspace
- Streams live output from multiple terminal windows side-by-side, allowing operators to monitor fleet progress in real time.
- Mid-Session Model Switching
- Swaps LLM reasoning engines on active sessions dynamically to adapt computing capability to changing task complexity.
- Automated Loop Schedulers
- Runs recurring cron-style prompts and autonomous overnight loops for continuous testing, documentation, and refactoring tasks.
- Self-Healing Watchdog
- Monitors active processes to automatically restart failed terminal sessions, prune memory contexts, and resume execution ledgers.
- Phone-First Remote Control
- Provides a responsive web interface optimized for mobile browsers, enabling full agent fleet supervision from anywhere.