AI Skill Hub 推荐使用:类体产常分 是一款优质的Agent工作流。AI 综合评分 7.5 分,在同类工具中表现稳健。如果你正在寻找可靠的Agent工作流解决方案,这是一个值得深入了解的选择。
导八产常源分果,导八产常源分果。导八产常源分果。导八产常源分果。
类体产常分 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
导八产常源分果,导八产常源分果。导八产常源分果。导八产常源分果。
类体产常分 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一:npm 全局安装 npm install -g toryo # 方式二:npx 直接运行(无需安装) npx toryo --help # 方式三:项目依赖安装 npm install toryo # 方式四:从源码运行 git clone https://github.com/JesseRWeigel/toryo cd toryo npm install npm start
# 命令行使用
toryo --help
# 基本用法
toryo [options] <input>
# Node.js 代码中使用
const toryo = require('toryo');
const result = await toryo.run(options);
console.log(result);
# toryo 配置说明 # 查看配置选项 toryo --config-example > config.yml # 常见配置项 # output_dir: ./output # log_level: info # workers: 4 # 环境变量(覆盖配置文件) export TORYO_CONFIG="/path/to/config.yml"
The intelligent agent orchestrator. Not just parallel agents — the full self-improving development loop.
棟梁 (toryo) — Japanese for "master builder" or "foreman." The toryo is the person who oversees the entire construction crew, assigns specialists to the right tasks, and ensures every piece meets quality standards before it stays in the structure.
Toryo chains multiple AI coding agents (Claude Code, Aider, Gemini CLI, Codex, Ollama) with spec-driven workflows, trust-based delegation, quality ratcheting, and a real-time dashboard.
npx @jweigel/toryo init # scaffold config + task specs
npx @jweigel/toryo run # start orchestration
Try instantly: npx @jweigel/toryo demo (no AI tools needed)
Documentation | Getting Started | Configuration | Bus Pattern | Contributing

---
Edit toryo.config.json:
{
"agents": {
"researcher": {
"adapter": "claude-code",
"strengths": ["research", "analysis"],
"timeout": 900
},
"coder": {
"adapter": "ollama",
"model": "qwen3.5:27b",
"strengths": ["code", "architecture"],
"timeout": 900
},
"reviewer": {
"adapter": "claude-code",
"strengths": ["review", "scoring"],
"timeout": 600
}
}
}
See examples/toryo.config.json for a complete example.
| Field | Type | Default | Description | |
|---|---|---|---|---|
name | string | — | Project name | |
agents | Record | — | Agent definitions (adapter, model, strengths, timeout) | |
tasks | string \ | TaskSpec[] | — | Path to specs dir or inline tasks |
ratchet.threshold | number | 6.0 | Minimum QA score to keep | |
ratchet.maxRetries | number | 1 | Ralph Loop max retries | |
ratchet.gitStrategy | string | "commit-revert" | "commit-revert", "branch-per-task", or "none" | |
delegation.initialTrust | number | 0.5 | Starting trust for new agents | |
delegation.scoreWindow | number | 50 | Rolling window for score averaging | |
outputDir | string | ".toryo" | Where to store results, metrics, artifacts | |
notifications.provider | string | "none" | "ntfy", "slack", "discord", "webhook", "none" |
Most multi-agent tools do one thing — run agents in parallel (Composio, AMUX) or define specs (Spec Kit). Toryo is the full loop: spec → delegate → execute → review → ratchet → improve.
| Feature | Toryo | Composio | AMUX | CrewAI | Spec Kit |
|---|---|---|---|---|---|
| Multi-agent orchestration | ✅ | ✅ | ✅ | ✅ | ❌ |
| Heterogeneous CLIs | ✅ 5+ adapters | ✅ 8 slots | ❌ Claude only | ❌ API only | ❌ |
| Spec-driven workflows | ✅ | ❌ | ❌ | ❌ | ✅ |
| Trust-based delegation | ✅ | ❌ | ❌ | ❌ | ❌ |
| Quality ratcheting | ✅ | ❌ | ❌ | ❌ | ❌ |
| Ralph Loop retries | ✅ | ❌ | ❌ | ❌ | ❌ |
| Auto-extraction | ✅ | ❌ | ❌ | ❌ | ❌ |
| results.tsv tracking | ✅ | ❌ | ❌ | ❌ | ❌ |
| Local model first | ✅ Ollama native | ❌ | ❌ | ❌ | ❌ |
| Real-time dashboard | ✅ | ✅ | ✅ | ❌ | ❌ |
导八产常分果、导八产常分果、导八产常分果、导八产常分果、导八产常分果、导八产常分果、导八产常分果、导八产常分果、导八产常分果、
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
总体来看,类体产常分 是一款质量良好的Agent工作流,在同类工具中具备一定竞争力。AI Skill Hub 将持续追踪其更新动态,建议收藏备用,结合自身场景选择合适时机引入使用。
| 原始名称 | toryo |
| Topics | workflowagentaiaiderautomationclaudetypescript |
| GitHub | https://github.com/JesseRWeigel/toryo |
| License | MIT |
| 语言 | TypeScript |
收录时间:2026-05-23 · 更新时间:2026-05-23 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端