GH Repository · huangruiteng
loopx
Long-horizon agent control plane for durable, governed work across Codex, Claude Code, and other harnesses.
- stars
- 5,854
- 30-day movement
- +8729/day
- Related entries
- 68
- Connections
- 2
pythonnodecodexloopxagent-appagent-harnessai-agentsagent-opsworkflow-automationlong-horizonloop-engineeringPythondsh-pluginagent-control-planelong-horizon-agents
loopx is a Python control plane for managing long-horizon AI agent work across harnesses such as Codex and Claude Code. It positions itself as a durable, governed layer that coordinates agent runs rather than being an agent itself.
Reach for it when you need a single governing layer to keep multi-step agent work durable and controlled across different agent harnesses.
Use it to
- Coordinate long-running agent tasks across Codex and Claude Code
- Add governance and durability to agent workflows
- Centralize agent operations in one control plane
- Automate multi-harness agent work
For Developers running long-horizon agents across multiple harnesses
- Role
- agent-app
- Language
- Python
- Licence
- Apache-2.0
- Forks
- 545
- Open issues
- 53
- Last push
- 2026-09-15
- Latest release
- v0.1.3 · 2026-07-02
- Skills shipped
- 8
topicsagent-control-planelong-horizon-agentsagent-opsworkflow-automationcodexclaude-code