Claude Skill

ruvnet/metaharness

Agent Harness Generator 是 AI 代理的元框架。使用 npx CLI、MCP 服务器、记忆、学习循环和见证签名发布,搭建品牌代理框架。兼容 Claude Code、Codex、pi.dev、Hermes、OpenClaw 和 RVM。

概览

Stars356
Forks34
语言TypeScript
最后更新2026-07-03
最近同步2026-07-03
前往 GitHub

仓库信息

拥有者ruvnet
仓库metaharness
完整名称ruvnet/metaharness
Repo ID1,268,611,404

安装这个 Skill

npx metaharness my-bot --template vertical:coding --host claude-code

Registry 信息

类型mcp_server
质量分80/100
验证状态readme_parsed
最近验证2026-06-22
平台
ClaudeMCPOpenClawCodex
能力
browsercode-reviewmemorysearchimageterminalworkflowagent-harnessagent-harness-generatoragent-scaffolding
识别文件
README.mddocsexamplespackage.json
配置键
URLPACKAGE_JSON
安装方式
  • npx metaharness my-bot --template vertical:coding --host claude-code
  • npx metaharness --wizard
  • npx metaharness --list
  • npx metaharness my-bot --template vertical:coding
  • git clone https://github.com/ruvnet/agent-harness-generator

项目简介

Agent Harness Generator 是 AI 代理的元框架,可让你搭建自己专注的品牌代理框架,配备专用 npx 命令行工具、MCP 服务器、记忆系统、学习循环以及见证签名发布机制。兼容 Claude Code、Codex、pi.dev、Hermes、OpenClaw 和 RVM(硬件隔离沙箱)。

英文描述

🛠️ The meta-harness for AI agents — scaffold your own focused, branded agent harness with its own npx CLI, MCP server, memory, learning loop, and witness-signed releases. Works with Claude Code, Codex, pi.dev, Hermes, OpenClaw, and RVM (hardware-isolated sandbox).

要点

  • 搭建品牌代理框架,配备自定义 npx 命令行工具
  • 内置 MCP 服务器,支持代理间通信
  • 持久化记忆与学习循环,实现持续改进
  • 见证签名发布,确保部署可验证
  • 兼容 Claude Code、Codex、pi.dev、Hermes、OpenClaw 和 RVM 沙箱

使用场景

  • 为特定 AI 工作流创建自定义代理框架
  • 构建基于 MCP 通信的多代理系统
  • 开发具备学习与记忆能力的自主代理
  • 通过见证签名部署可验证的代理发布
  • 集成 Claude Code、Codex 或 Hermes,增强自动化能力

README 摘要

<div align="center"> # MetaHarness ### Mint a custom AI agent harness from any repo. `npx metaharness` · [open the Studio →](https://ruvnet.github.io/agent-harness-generator/) <sub>(Repo: `ruvnet/agent-harness-generator` · CLI: `metaharness` · Library: `@ruvnet/agent-harness-generator`)</sub> [![Open the Studio](https://img.shields.io/badge/Studio-open_in_browser_↗-7c5cff?style=for-the-badge&logo=githubpages&logoColor=white)](https://ruvnet.github.io/agent-harness-generator/) [![User guide](https://img.shields.io/badge/User_guide-plain_language-22c55e?style=for-the-badge)](docs/USERGUIDE.md) [![Tests — 568 passing](https://img.shields.io/badge/tests-568%20passing-22c55e?style=for-the-badge)](docs/ARCHITECTURE.md) [![npm publish smoke](https://github.com/ruvnet/agent-harness-generator/actions/workflows/published-smoke.yml/badge.svg)](https://github.com/ruvnet/agent-harness-generator/actions/workflows/published-smoke.yml) [![License MIT](https://img.shields.io/badge/License-MIT-yellow?style=for-the-badge)](LICENSE) [![Agent Harness Studio](docs/web-ui/screenshot-desktop.png)](https://ruvnet.github.io/agent-harness-generator/) </div> --- ## What this is **Every serious repo deserves its own agent.** A repo-aware CLI, a repo-aware coding agent, a local MCP server, memory scoped to the project, skills generated from the actual file layout, governance policy, release verification, witness-signed provenance. `metaharness` mints those, on demand, from a GitHub URL or a blank slate. **It is not another agent framework. It is a factory for agent frameworks.** The model is replaceable. The harness is the product. ## What it gives you In under 60 seconds, in your browser, with nothing leaving your machine: - A custom AI agent harness for your repo (or any repo) - Rec

话题

探索更多

数据来自 GitHub,同步时间:2026-07-03