Launching today
Hooman
Own your AI agent stack — local, open source and extensible
8 followers
Own your AI agent stack — local, open source and extensible
8 followers
Hooman is an open-source, local-first AI agent runtime that works across the terminal, VS Code, ACP-compatible editors and event-driven workflows. Use your preferred models and inference endpoints, including local llama.cpp and MLX, or bring your own API keys. Hooman includes agent, plan, ask and design modes, MCP integrations, reusable skills, subagents, approval controls and persistent project-scoped sessions.
Would love to see a built-in TUI dashboard for monitoring active sessions, token usage, and subagent status at a glance. Something like a tmux-style interface where I can attach to a session, see what the agent is currently thinking, intervene mid-run, and pause or kill it without dropping context. Would make managing long-running local workflows way less opaque.
Local-first is a great call, would love to see a built-in cost and token usage dashboard per session and provider so I can keep tabs on API spend without wiring up my own logging.