AI Skill Hub 强烈推荐:连接洋葱 是一款优质的Agent工作流。已获得 1.2k 颗 GitHub Star,AI 综合评分 8.0 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。
连接洋葱 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。
连接洋葱 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:pip 安装(推荐)
pip install connectonion
# 方式二:虚拟环境安装(推荐生产环境)
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install connectonion
# 方式三:从源码安装(获取最新功能)
git clone https://github.com/openonion/connectonion
cd connectonion
pip install -e .
# 验证安装
python -c "import connectonion; print('安装成功')"
# 命令行使用
connectonion --help
# 基本用法
connectonion input_file -o output_file
# Python 代码中调用
import connectonion
# 示例
result = connectonion.process("input")
print(result)
# connectonion 配置文件示例(config.yml) app: name: "connectonion" debug: false log_level: "INFO" # 运行时指定配置文件 connectonion --config config.yml # 或通过环境变量配置 export CONNECTONION_API_KEY="your-key" export CONNECTONION_OUTPUT_DIR="./output"
A simple, elegant open-source framework for production-ready AI agents
📚 Documentation • 💬 Discord • ⭐ Star Us
</div>
---
## 🌟 Philosophy: "Keep simple things simple, make complicated things possible" This is the core principle that drives every design decision in ConnectOnion.
agent = Agent("researcher", tools=[search], plugins=[ re_act, # Reflect + plan after each tool call auto_compact, # Auto-compress context at 90% capacity subagents, # Spawn sub-agents with independent tools and prompts ulw, # Ultra Light Work — fully autonomous mode ]) ```
These plugins mirror Claude Code's internal capabilities — auto_compact, subagents, ulw directly correspond to Claude Code's context compression, sub-agent spawning, and autonomous work mode. ConnectOnion makes these capabilities available to any agent you build.
Hooks: after_user_input, before_iteration, before_llm, after_llm, before_tools, before_each_tool, after_each_tool, after_tools, on_error, after_iteration, on_stop_signal, on_complete
Plugins are just lists of event handlers — visible, modifiable, co copy-able.
| Feature | Description |
|---|---|
| Built-in AI Programmer | co ai - AI coding assistant |
| Built-in Frontend & Backend | chat.openonion.ai ready-to-use |
| Ready-to-Use Tools | Import without schema writing |
| Approval System | Dangerous ops auto-trigger approval |
| Skills System | Claude Code compatible, auto-discovery |
| 12 Lifecycle Hooks | Inject logic at any point |
| Plugin System | re_act, auto_compact, subagents, ulw |
| Multi-Agent Trust | Fast rules, zero token cost |
```
Plugin-based: turn it off, customize it, or replace it entirely.
Python 3.10+
pip install connectonion
from connectonion import host
host(agent) # HTTP + P2P relay
```python import os from connectonion import Agent
- Always acknowledge the customer's concern first - Look for root causes, not just symptoms - Provide clear, actionable solutions ```
You can still use the traditional Tool class approach, but the new functional approach is much simpler:
pip install connectonion
os.environ["OPENAI_API_KEY"] = "your-api-key-here"
ConnectOnion CLI provides templates to get you started quickly:
```bash
Set your API key via environment variable:
export OPENAI_API_KEY="your-api-key-here"
Or pass directly to agent:
agent = Agent(name="test", api_key="your-api-key-here")
Inject logic at any point in the agent execution cycle:
```python from connectonion import Agent, after_tools, llm_do from connectonion.useful_plugins import re_act, eval, auto_compact, subagents, ulw
Package reusable capabilities as plugins and use them across multiple agents:
```python from connectonion import Agent, after_tools, llm_do
def add_reflection(agent): trace = agent.current_session['trace'][-1] if trace['type'] == 'tool_execution' and trace['status'] == 'success': result = trace['result'] reflection = llm_do( f"Result: {result[:200]}\n\nWhat did we learn?", system_prompt="Be concise.", temperature=0.3 ) agent.current_session['messages'].append({ 'role': 'assistant', 'content': f"🤔 {reflection}" })
reflection = [after_tools(add_reflection)] # after_tools fires once after all tools
agent = Agent( name="support_agent", system_prompt="prompts/customer_support.md" # Automatically loads file content )
Plugins are lists of lifecycle hooks that inject logic at any point in the agent execution cycle. Built-in plugins: - re_act: Reflect + plan after each tool call - auto_compact: Auto-compress context at 90% capacity - subagents: Spawn sub-agents with independent tools - ulw: Ultra Light Work — fully autonomous mode
from connectonion.useful_plugins import re_act, subagents
agent = Agent("researcher", tools=[search], plugins=[re_act, subagents])
| Plugin | Description | Claude Code Equivalent |
|---|---|---|
| re_act | Reflect + plan after each tool | - |
| auto_compact | Auto-compress context at 90% | Context compression |
| subagents | Spawn sub-agents | Sub-agent spawning |
| ulw | Ultra Light Work autonomous | Autonomous mode |
高质量的开源AI工作流框架
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。
✅ Apache 2.0 — 宽松开源协议,可商用,需保留版权声明和 NOTICE 文件,含专利授权条款。
总体来看,连接洋葱 是一款质量优秀的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 | connectonion |
| Topics | AI工作流智能体 |
| GitHub | https://github.com/openonion/connectonion |
| License | Apache-2.0 |
| 语言 | Python |
收录时间:2026-06-04 · 更新时间:2026-06-04 · License:Apache-2.0 · 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