Issue with OpenAIAssistant as a node in graph

[vsnakul]

I have an open ai assistant created and have been using it directly via openai python SDK and have been working smoothly....Recently i had a requirement in which i had to interact with multiple assistants to get to a final state.
I was trying out the langgraph library to orchestrate this flow with each assistant and function call as a node in the graph.

Please see the code below

`const StateAnnotation = Annotation.Root({
    reducer: messagesStateReducer,
  }),
});
const baseAgentTools = [new CreateCaseFn()];
const toolNode = new ToolNode(baseAgentTools);
const baseAgentNode = new OpenAIAssistantRunnable({
  assistantId:process.env.ASST,
  asAgent: true,
});
const agentExecutor = AgentExecutor.fromAgentAndTools({
  agent: baseAgentNode,
  tools: baseAgentTools,
  maxIterations: 1,
  verbose: true,
});

async function callBaseAgent(state: typeof StateAnnotation.State) {
  const messages = state.messages;
  const response = await agentExecutor.invoke(messages);
  return { messages: [response] };
}
function shouldContinue(state: typeof StateAnnotation.State) {
  const messages = state.messages;
  const lastMessage = messages[messages.length - 1] as AIMessage;
  if (lastMessage.tool_calls?.length) {
    return 'create_case_fn';
  }
  return '__end__';
}
const workflow = new StateGraph(StateAnnotation)
  .addNode('base_agent', callBaseAgent)
  .addNode('create_case_fn', toolNode)
  .addEdge('__start__', 'base_agent')
  .addConditionalEdges('base_agent', shouldContinue)
  .addEdge('base_agent', 'create_case_fn');

const checkpointer = new MemorySaver();

const app = workflow.compile({ checkpointer });

const start = async () => {
  const finalState = await app.invoke(
    { messages: [new HumanMessage('what is the weather in sf')] },
    { configurable: { thread_id: '42' } },
  );
  console.log(finalState.messages[finalState.messages.length - 1].content);
};
start();

See the below logs

(node:31200) [DEP0040] DeprecationWarning: The punycodemodule is deprecated. Please use a userland alternative instead. (Usenode --trace-deprecation ... to show where the warning was created) [chain/start] [1:chain:AgentExecutor] Entering Chain run with input: { "0": { "lc": 1, "type": "constructor", "id": [ "langchain_core", "messages", "HumanMessage" ], "kwargs": { "content": "what is the weather in sf", "additional_kwargs": {}, "response_metadata": {} } } } [chain/error] [1:chain:AgentExecutor] [495ms] Chain run errored with error: "400 Missing required parameter: 'thread.messages[0].content'.\n\nError: 400 Missing required parameter: 'thread.messages[0].content'.\n at Function.generate (C:\\Projects\\proj\\prof-consumer\\node_modules\\@langchain\\openai\\node_modules\\openai\\src\\error.ts:72:14)\n at OpenAI.makeStatusError (C:\\Projects\\proj\\proj-consumer\\node_modules\\@langchain\\openai\\node_modules\\openai\\src\\core.ts:435:21)\n at OpenAI.makeRequest (C:\\Projects\\proj\\proj-consumer\\node_modules\\@langchain\\openai\\node_modules\\openai\\src\\core.ts:499:24)\n at processTicksAndRejections (node:internal/process/task_queues:95:5)\n at async OpenAIAssistantRunnable.invoke (C:\\Projects\\proj\\proj-consumer\\node_modules\\langchain\\dist\\experimental\\openai_assistant\\index.cjs:91:19)\n at async OpenAIAssistantRunnable._streamIterator (C:\\Projects\\proj\\proj-consumer\\node_modules\\@langchain\\core\\dist\\runnables\\base.cjs:165:9)" C:\Projects\proj\proj-consumer\node_modules\@langchain\openai\node_modules\openai\src\error.ts:72 return new BadRequestError(status, error, message, headers); ^ BadRequestError: 400 Missing required parameter: 'thread.messages[0].content'. at Function.generate (C:\Projects\proj\proj-consumer\node_modules\@langchain\openai\node_modules\openai\src\error.ts:72:14) at OpenAI.makeStatusError (C:\Projects\proj\proj-consumer\node_modules\@langchain\openai\node_modules\openai\src\core.ts:435:21) at OpenAI.makeRequest (C:\Projects\proj\proj-consumer\node_modules\@langchain\openai\node_modules\openai\src\core.ts:499:24) at processTicksAndRejections (node:internal/process/task_queues:95:5) at async OpenAIAssistantRunnable.invoke (C:\Projects\proj\proj-consumer\node_modules\langchain\dist\experimental\openai_assistant\index.cjs:91:19) at async OpenAIAssistantRunnable._streamIterator (C:\Projects\proj\proj-consumer\node_modules\@langchain\core\dist\runnables\base.cjs:165:9)

Is there something wrong with the way i am invoking my assistant ?
Any help is greately appreciated