← all releases
AMAP (Alibaba)·Sep 14, 2026·v0.1.7·open source

LongHorizon-Harness

LongHorizon-Harness is a loop engineering framework designed to enable AI agents to perform complex, long-horizon tasks across desktop applications and CLI environments. It acts as a durable execution layer that manages task state, verifies results independently, and recovers from failures, allowing existing agents like Claude Code, Codex, and DeepSeek Harness to operate autonomously for extended periods.

AI AgentsComputer UseAutomationLoop Engineering
visit website github ↗
overview

LongHorizon-Harness is a loop engineering framework designed to enable AI agents to perform complex, long-horizon tasks across desktop applications and CLI environments. It acts as a durable execution layer that manages task state, verifies results independently, and recovers from failures, allowing existing agents like Claude Code, Codex, and DeepSeek Harness to operate autonomously for extended periods.

key features
  • 01Manage-Execute-Audit (MEA) loop for reliable task state management
  • 02Independent auditing of agent actions to ensure ground truth
  • 03Seamless integration with Claude Code, Codex, OpenCode, and DeepSeek Harness
  • 04Support for both GUI desktop applications and CLI terminal workflows
  • 05React/FastAPI-based web workbench for real-time task monitoring and human intervention
  • 06Automatic recovery and checkpointing for long-running, complex tasks
related productsbrowse all →
MiroThinkeropen sourceMiroThinker is an open-source search agent model developed by MiroMindAI, designed for tool-augmented reasoning and real-world information seeking. It aims to match the deep research capabilities of leading AI models like OpenAI's Deep Research and Google's Gemini Deep Research.Letta Code SDKopen sourceThe Letta Code SDK is a software development kit that enables developers to build deeply personalized agents with persistent memory that learn over time. It serves as the interface to Letta Code, facilitating the creation of stateful agents capable of continuous learning and improvement.OB-1open sourceOB-1 is a self-improving coding agent developed by OpenBlock Labs, designed to autonomously handle the full development lifecycle, from project management to pull requests. It integrates seamlessly into existing workflows, enhancing productivity and code quality.