Hello , I am student trying to learn langgraph. I am facing a problem and can’t figure out why i am getting error while using truecall extractor can you look at my code and tell me what the problem is.
those lines:
prompt = store_memory_prompt.format(memory=str(memories))
response = extractor.invoke({"messages":[SystemMessage(content=store_memory_prompt)]+state["messages"],"existing":existing_memory})
Should likely be like:
prompt = store_memory_prompt.format(memory=str(memories))
response = extractor.invoke({"messages":[SystemMessage(content=prompt)]+state["messages"],"existing":existing_memory})
content=store_memory_prompt → content=prompt
It still fails, and gives validation error. I have researched and I think the problem is that gemini model doesn’t work well with trustcall.