Zeeland
## Introduction Sometime, we need some custom prompt to drive our LLM. Currently, there is no convenient mechanism to handle it in chatGPTBox. **We can also see other relevant demands:**...
I have already used langgraph in some production projects. When dealing with particularly complex logic, it is easy to misspell node names by simply using string to declare node names....
## 🚀 Feature Request Currently, we cannot enable stream if setting output schema. Eg: ```python from typing import List import promptulate as pne from pydantic import BaseModel, Field class LLMResponse(BaseModel):...
## 🚀 Feature Request We hope use some tiny [user case ](https://undertone0809.github.io/promptulate/#/modules/usercases/intro?id=user-cases) to help development startup. An idea is we can build a AI websearch website for user. We can...
## 🚀 Feature Request Add the retry mechanism to increase the robustness of the system because LLM output is unstable and sometimes not output in the specified format if formatted....
## 🚀 Feature Request Add pne.chat() use openai provider to proxy some specified model, eg: ```python import promptulate as pne pne.chat(messages="hello", model="gpt-4-turbo") ``` If developer want to use openai provider...
remove ChatOpenAI usage, pne.LLMFactory.build(...) is recommended. docs: https://undertone0809.github.io/promptulate/#/modules/llm/llm-factory-usage?id=llmfactory Click here to review the code @Hizeros @ruanrongman 
- Add an websearch app by streamlit pne.ToolAgent and relevant tools. - Add into docs https://undertone0809.github.io/promptulate/#/use_cases/intro?id=use-cases
## 🚀 Feature Request [Agent-protocol](https://agentprotocol.ai/) is a universal agent communication protocol. We can implement agent-protocol to build a complex Agent. I hope all of our Agent can implement Agent-protocol