Local-first agent OS
Define an AI agent in YAML, run it, and it persists. No Docker, no cloud. Agents route between multiple LLM backends (Claude, Codex, LM Studio) and collaborate through shared "rooms." They synthesize their own tools when needed.
Every agent framework I've tried either requires a PhD in configuration or ships your data to someone else's cloud. Hive is the alternative: your machine, your models, your agents.
Python, Typer CLI, YAML profiles, JSONL persistence + SQLite state, multi-model routing, agent-to-agent communication via rooms.
Phase 1: daemon loop, autonomy system, multi-model routing. Then: tool synthesis, suffering system (stressors drive agent behavior change), and integration with Mutter for voice-triggered agents.