Persistent memory for artificial intelligence agents
Loopx is a lightweight state kernel developed for long-term artificial intelligence agent teams. It works compatible with coding agents (Codex, Claude Code) and standardizes processes such as task tracking, quota management and verifiable work transfer.
Updates
- September 16, 2026: Stars 5,853 → 5,864, latest release v1.0.5 (September 15, 2026).
- September 15, 2026: Stars 5,793 → 5,853, latest release v1.0.4 (September 15, 2026).
- September 11, 2026: Stars 5,751 → 5,793, latest release v1.0.3 (September 11, 2026).
- September 9, 2026: Stars 5,726 → 5,751, latest release v1.0.2 (September 9, 2026).
What you get
- Standardizes target tracking and quota management in long-term missions.
- It enables verifiable handover of work between different AI agents.
- Stores task history and evidence in an organized structure.
Installation
curl -fsSL https://raw.githubusercontent.com/huangruiteng/loopx/main/scripts/install-from-github.sh | bash
export PATH="$HOME/.local/bin:$PATH"
loopx doctorRunning it
cd /path/to/your-project
loopx connect
loopx statusIf you don't write code
We will work on a long-term project with you. I want to manage our tasks, goals, and progress logs with the LoopX state bean. As you work, check the current status at each step, record evidence of completed work, and stay within quota limits by determining the next todo (todo). Leave a verifiable trail through LoopX when delegating tasks or moving to a new phase.
Related dictionary terms
Links
TreScout did not build this tool · we found it in GitHub trends and wrote it up. This page describes the repository as of 2026-08-06: The star count and our text belong to that day, the repository may have changed since. Check the repository link for the current state. This page was machine-translated from the Turkish original · the Turkish version prevails.