LoopX
Local-first control plane for long-running coding and research agents
What it is
LoopX keeps AI agents such as Codex and Claude Code working toward durable goals across sessions, tracking todos, handoffs, decisions and evidence in a local workspace. It adds a browser and desktop dashboard for steering live turns, reviewing deliverables, and analyzing agent trajectories with citations and cost readback. Developers and research teams running multi-day agent work use it to recover state after restarts and coordinate several agents on one goal.