huangruiteng/loopx

99 stars · Last commit 2026-07-13

Lightweight loop engineering state kernel for long-running AI agent teams. Agent-loop agnostic across Codex, Claude Code, and other coding agents, with durable goals, quota-aware auto-wake, executable todos, evidence logs, and verifiable handoffs.

README preview

<div align="center">

<img src="docs/assets/loopx-social-preview.png" alt="LoopX loop engineering social preview banner" width="480">

**Loop engineering for long-running AI agents and peer agent teams.**

<sub>A lightweight state kernel and agent-agnostic local control plane for
Codex, Claude Code, Cursor, and other runtimes: objectives, gates, todos,
quota, scheduler hints, evidence, and typed continuation in one reviewable loop.</sub>

[![License](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
[![Python](https://img.shields.io/badge/python-3.11%2B-blue.svg)](pyproject.toml)
[![Local first](https://img.shields.io/badge/control--plane-local--first-brightgreen.svg)](docs/public-private-boundary.md)
[![Loop Agents](https://img.shields.io/badge/status-loop%20agents%20early-orange.svg)](docs/product/release-readiness.md)

**把会干活的 Agent，接成可管理、可复盘、可持续改进的数字员工。**

</div>

---

View full repository on GitHub →