Problem solved

diri solves the challenge of coordinating multiple AI-powered coding assistants concurrently across isolated git worktrees and remote environments. It streamlines complex development tasks by enabling parallelism, reducing context-switching, and providing a unified monitoring interface, thus improving developer productivity and workflow efficiency.

Use cases

• Running multiple AI coding agents in parallel on different git branches or worktrees
• Orchestrating coding tasks across local and remote hosts from a single interface
• Monitoring live status of AI-assisted coding sessions in real time
• Persisting active agent sessions across system restarts to continue work seamlessly
• Quickly toggling between agents and sessions via a menu-bar control
• Combining outputs from different AI models (Claude, Gemini, etc.) for complex refactoring

Audience

• Developers
• ML Engineers
• Data Scientists

Pros

• Unified interface for orchestrating multiple AI coding tools and shells
• Supports parallel execution across git worktrees and remote hosts
• Session persistence ensures continuity across restarts
• Convenient menu-bar integration for quick access and management
• Open source and free to use

Cons

• Only available for macOS
• Requires separate installation and configuration of supported AI tools
• Managing many parallel agents may introduce cognitive overhead
• Dependent on third-party AI services and their availability
• Documentation and community support may be limited as an early-stage project