👐 Create agent tools

The easiest way to define custom tools for agents in LangChain is using the @tool decorator. The decorator makes tools out of functions by using the function name as the tool name by default, and the function's docstring as the tool's description.

We want the MongoDB learning assistant to have access to the following tools:

• get_information_for_question_answering: Uses vector search to retrieve information to answer questions

• get_article_content_for_summarization: Gets the content of articles for summarization

Fill in any <CODE_BLOCK_N> placeholders and run the cells under the Step 5: Create agent tools section in the notebook to create tools for the agent to use.

The answers for code blocks in this section are as follows:

CODE_BLOCK_3

Answer

embedding_model.encode(text)

CODE_BLOCK_4

Answer

get_embedding(user_query)

CODE_BLOCK_5

Answer

[
    {
        "$vectorSearch": {
            "index": VS_INDEX_NAME,
            "path": "embedding",
            "queryVector": query_embedding,
            "numCandidates": 150,
            "limit": 5,
        }
    },
    {
        "$project": {
            "_id": 0,
            "body": 1,
            "score": {"$meta": "vectorSearchScore"},
        }
    },
]

CODE_BLOCK_6

Answer

vs_collection.aggregate(pipeline)

CODE_BLOCK_7

Answer

mongodb_client[DB_NAME][FULL_COLLECTION_NAME]

CODE_BLOCK_8

Answer

{"title": user_query}

CODE_BLOCK_9

Answer

{"_id": 0, "body": 1}

CODE_BLOCK_10

Answer

full_collection.find_one(query, projection)