LLM-201: OpenHands与LLM交互链路分析

一、核心交互链路架构

二、详细流程分解

1. 前端交互层
   React组件通过React Query发起API请求：

// OpenHands/frontend/src/components/ChatInput.tsx
const { trigger } = useSWRMutation('/api/chat', sendMessage);async function sendMessage(url: string, { arg }: { arg: string }) {return axios.post(url, {session_id: sessionId,message: arg});
}

2. API路由层
   FastAPI处理请求并创建会话：

# OpenHands/openhands/server/routes/chat.py
@app.post("/chat")
async def chat_endpoint(request: ChatRequest):session = AgentSessionManager.get_session(request.session_id)await session.start()await session.process_event(MessageAction(content=request.message))return EventStreamResponse(session.event_stream)

3. Agent控制层
   AgentController主循环处理事件：

# OpenHands/openhands/controller/agent_controller.py
async def _execute_step(self):messages = self.conversation_memory.process_events(...)llm_response = await self.llm.acompletion(messages)actions = self.agent.response_to_actions(llm_response)for action in actions:await self._handle_action(action)

4. LLM交互层
   通过LiteLLM集成多模型：

# OpenHands/openhands/llm/llm.pyasync def acompletion(self, messages: list[Message]) -> ModelResponse:return await litellm.acompletion(model=self.config.model_name,messages=convert_to_oa_messages(messages),tools=self.tool_schema)

5. 工具执行层
   文件编辑工具示例：

# OpenHands/openhands/tools/file_edit.py
class FileEditTool(BaseTool):async def execute(self, params: dict) -> FileEditObservation:with open(params['filepath'], 'w') as f:f.write(params['content'])return FileEditObservation(content=f"Updated {params['filepath']}")

三、典型交互示例

1. 用户请求
   前端发送：POST /chat {"message": "修改README.md第5行"}

2. 链路处理

3. 结果反馈

前端接收SSE事件：

{"type": "observation","data": {"content": "Successfully updated README.md","type": "file_edit"}
}

四、关键技术特性

1. 实时事件流：通过Server-Sent Events实现低延迟更新
2. 上下文管理：ConversationMemory维护500轮对话上下文
3. 错误恢复：_react_to_exception方法实现异常自动处理
4. 多模型支持：LLM配置支持30+商业/开源模型接入

五、参考

1. OpenHands document
2. OpenHands on Github