[HANDS-ON BUG] Unit 3| Creating a RAG Tool for Guest Stories | Step 2 | langgraph

For the AI Agent Langgraph version the code needs update.

Extracting text function "extract_text" tries to return "\n\n".join([doc.text for doc in results[:3]]), and it gives an error in tool call:

{'tools': {'messages': [ToolMessage(content='Error: AttributeError("\'Document\' object has no attribute \'text\'")\n Please fix your mistakes.', name='guest_info_retriever', id='c34e52a8-3ba2-437b-a0f3-dadfe1fa99e1', tool_call_id='0', status='error')]}}

To allow the tool to return data, next change will help (doc.text replace by doc.page_content):

return "\n\n".join([doc.text for doc in results[:3]]) -> return "\n\n".join([doc.page_content for doc in results[:3]])

Start with the extract_text function in Unit 3, Creating a RAG Tool for Guest Stories, Step 2, and reproduce the guest_info_retriever tool call shown in the issue. Verify that the tool returns the retrieved guest-story data without the Document attribute error.

Written by the indexing model from the issue text.