Show HN: Lazyagent – a local TUI for watching what your coding agents are doing (github.com)

Lazyagent has emerged as a new terminal-based TUI (Text User Interface) application that allows developers to monitor the runtime activities of AI agents like Claude, Codex, and OpenCode. The tool captures various runtime events, stores them in an SQLite database, and presents them in an organized interface, making it easier for users to track project sessions, agents, subagents, and outputs without sifting through complex raw hook data. The utility enhances day-to-day observability, empowering users to quickly understand which session belongs to which project and the status of different agents and tools in play. This development is significant for the AI/ML community as it strengthens the debugging and tracking capabilities of AI projects, particularly in contexts where multiple agents are in operation simultaneously. By streamlining the visualization of events and interactions, Lazyagent enables better project management and enhances collaboration among developers. The app boasts various installation options, supports multiple event hooks for each AI agent, and features a user-friendly interface with functions for data refresh, project grouping, and event filtering, making it a powerful tool for both individual developers and teams engaged in AI/ML workflows.