https://github.com/HKUDS/OpenHarnessOpenHarness提供核心的轻量级智能体基础设施工具使用、技能、记忆和多智能体协作。ohmo是一个基于 OpenHarness 构建的个人 AI 智能体——它不是一个普通的聊天机器人而是一个能在长时间会话中真正为你工作的助手。你可以在飞书 / Slack / Telegram / Discord 中与 ohmo 聊天它会自动创建分支、编写代码、运行测试并提交 PR。ohmo 运行在你现有的 Claude Code 或 Codex 订阅上——无需额外的 API 密钥。加入社区为开放的智能体开发贡献Harness。OpenHarnessdelivers core lightweight agent infrastructure: tool-use, skills, memory, and multi-agent coordination.ohmois a personal AI agent built on OpenHarness — not another chatbot, but an assistant that actually works for you over long sessions. Chat with ohmo in Feishu / Slack / Telegram / Discord, and it forks branches, writes code, runs tests, and opens PRs on its own. ohmo runs on your existing Claude Code or Codex subscription — no extra API key needed.Join the community: contributeHarnessfor open agent development.Agent Harness智能体套件是围绕大语言模型LLM构建的完整基础设施旨在将其转化为功能完备的智能体。模型提供智能套件则提供手、眼、记忆和安全边界。OpenHarness是一个开源的 Python 实现专为研究者、构建者和社区设计理解生产级 AI 智能体的内部工作原理实验前沿工具、技能和智能体协作模式扩展套件添加自定义插件、提供程序和领域知识构建基于成熟架构的专用智能体https://github.com/HKUDS/OpenHarness.gitAnAgent Harnessis the complete infrastructure that wraps around an LLM to make it a functional agent. The model provides intelligence; the harness provideshands, eyes, memory, and safety boundaries.OpenHarness is an open-source Python implementation designed forresearchers, builders, and the community:Understandhow production AI agents work under the hoodExperimentwith cutting-edge tools, skills, and agent coordination patternsExtendthe harness with custom plugins, providers, and domain knowledgeBuildspecialized agents on top of proven architecture---openharness/engine/ # Agent Loop — query → stream → tool-call → looptools/ # 43 Tools — file I/O, shell, search, web, MCPskills/ # Knowledge — on-demand skill loading (.md files)plugins/ # Extensions — commands, hooks, agents, MCP serverspermissions/ # ️ Safety — multi-level modes, path rules, command denyhooks/ # ⚡ Lifecycle — PreToolUse/PostToolUse event hookscommands/ # 54 Commands — /help, /commit, /plan, /resume, ...mcp/ # MCP — Model Context Protocol clientmemory/ # Memory — persistent cross-session knowledgetasks/ # Tasks — background task managementcoordinator/ # Multi-Agent — subagent spawning, team coordinationprompts/ # Context — system prompt assembly, CLAUDE.md, skillsconfig/ # ⚙️ Settings — multi-layer config, migrationsui/ # ️ React TUI — backend protocol frontend
https://github.com/HKUDS/OpenHarnessOpenHarness提供核心的轻量级智能体基础设施工具使用、技能、记忆和多智能体协作。ohmo是一个基于 OpenHarness 构建的个人 AI 智能体——它不是一个普通的聊天机器人而是一个能在长时间会话中真正为你工作的助手。你可以在飞书 / Slack / Telegram / Discord 中与 ohmo 聊天它会自动创建分支、编写代码、运行测试并提交 PR。ohmo 运行在你现有的 Claude Code 或 Codex 订阅上——无需额外的 API 密钥。加入社区为开放的智能体开发贡献Harness。OpenHarnessdelivers core lightweight agent infrastructure: tool-use, skills, memory, and multi-agent coordination.ohmois a personal AI agent built on OpenHarness — not another chatbot, but an assistant that actually works for you over long sessions. Chat with ohmo in Feishu / Slack / Telegram / Discord, and it forks branches, writes code, runs tests, and opens PRs on its own. ohmo runs on your existing Claude Code or Codex subscription — no extra API key needed.Join the community: contributeHarnessfor open agent development.Agent Harness智能体套件是围绕大语言模型LLM构建的完整基础设施旨在将其转化为功能完备的智能体。模型提供智能套件则提供手、眼、记忆和安全边界。OpenHarness是一个开源的 Python 实现专为研究者、构建者和社区设计理解生产级 AI 智能体的内部工作原理实验前沿工具、技能和智能体协作模式扩展套件添加自定义插件、提供程序和领域知识构建基于成熟架构的专用智能体https://github.com/HKUDS/OpenHarness.gitAnAgent Harnessis the complete infrastructure that wraps around an LLM to make it a functional agent. The model provides intelligence; the harness provideshands, eyes, memory, and safety boundaries.OpenHarness is an open-source Python implementation designed forresearchers, builders, and the community:Understandhow production AI agents work under the hoodExperimentwith cutting-edge tools, skills, and agent coordination patternsExtendthe harness with custom plugins, providers, and domain knowledgeBuildspecialized agents on top of proven architecture---openharness/engine/ # Agent Loop — query → stream → tool-call → looptools/ # 43 Tools — file I/O, shell, search, web, MCPskills/ # Knowledge — on-demand skill loading (.md files)plugins/ # Extensions — commands, hooks, agents, MCP serverspermissions/ # ️ Safety — multi-level modes, path rules, command denyhooks/ # ⚡ Lifecycle — PreToolUse/PostToolUse event hookscommands/ # 54 Commands — /help, /commit, /plan, /resume, ...mcp/ # MCP — Model Context Protocol clientmemory/ # Memory — persistent cross-session knowledgetasks/ # Tasks — background task managementcoordinator/ # Multi-Agent — subagent spawning, team coordinationprompts/ # Context — system prompt assembly, CLAUDE.md, skillsconfig/ # ⚙️ Settings — multi-layer config, migrationsui/ # ️ React TUI — backend protocol frontend
)
