AMAP-ML/LongHorizon-Harness

  • 类型:github
  • 标识:AMAP-ML/LongHorizon-Harness
  • 链接:https://github.com/AMAP-ML/LongHorizon-Harness
  • 主题:agent, llm-infra
  • 主分类:evaluation
  • 形态:tool
  • 分类:trending
  • Stars:571
  • 周增:+0
  • 语言:Python
  • 许可:MIT
  • 最近提交:2026-08-11
  • 简介:The long-horizon computer-use harness. Run AI agents across desktop apps and the CLI for extended periods while preserving task state and making reliable progress on complex workflows. Features fresh-context execution, durable verified state, independent auditing, recoverable progress, and native Claude Code / Codex / OpenClaw integration.
  • 上次采集:2026-08-11
  • 首次采集:2026-08-11
  • 待LLM分类:否
  • 成熟度:research
  • 简介中文:长时程计算机使用 harness。让 AI Agent 在桌面应用与 CLI 中长时间运行,同时保持任务状态并在复杂工作流中可靠推进。具备全新上下文执行、持久化已验证状态、独立审计、可恢复进度以及原生 Claude Code / Codex / OpenClaw 集成。
  • 来源文件
  • [GitHub Search]