# Run long-horizon computer-use agent loops with LongHorizon-Harness

> Use LongHorizon-Harness to keep Claude Code, Codex, DeepSeek Harness, or custom agent backends working through long GUI and CLI tasks with verified state, role separation, checkpoints, and recovery loops.

- Skill: `agentskillexchange/run-long-horizon-computer-use-agent-loops-with-longhorizon-h` (Agent Skill)
- Install (CLI): `npx skillmds@latest add agentskillexchange/run-long-horizon-computer-use-agent-loops-with-longhorizon-h`
- Raw SKILL.md: https://api.skillmd.com/api/skills/agentskillexchange/run-long-horizon-computer-use-agent-loops-with-longhorizon-h/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: agentskillexchange (https://skillmd.com/u/agentskillexchange)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/agentskillexchange/run-long-horizon-computer-use-agent-loops-with-longhorizon-h

---


# Run long-horizon computer-use agent loops with LongHorizon-Harness

Use LongHorizon-Harness to keep Claude Code, Codex, DeepSeek Harness, or custom agent backends working through long GUI and CLI tasks with verified state, role separation, checkpoints, and recovery loops.

## Prerequisites

Python >= 3.10; uv or pip; one agent runtime on PATH such as codex, claude, or dsh; Node.js 20+ for computer-use plugins when GUI tasks are needed

## Installation

Install or set up from the source-backed instructions:

uv tool install lh-harness # or: pip install lh-harness cd /path/to/your/project lh-harness init lh-harness web --workspace-root . # or: lh-harness run --task "Inspect the current directory and summarize its files." --agent codex

- Source: https://github.com/AMAP-ML/LongHorizon-Harness

## Documentation

- https://lh-harness.pages.dev

## Source

- [Agent Skill Exchange](https://agentskillexchange.com/skills/run-long-horizon-computer-use-agent-loops-with-longhorizon-harness/)