Chat Sessions with Amazon DynamoDB

Griptape provides Conversation Memory as a means of persisting conversation context across multiple Structure runs. If you provide it with a suitable Driver, the memory of the previous conversation can be preserved between run of a Structure, giving it additional context for how to respond. While we can use the LocalConversationMemoryDriver to store the conversation history in a local file, in production use-cases we may want to store in a proper database.

In this example, we will show you how to use the AmazonDynamoDbConversationMemoryDriver to persist the memory in an Amazon DynamoDB table. Please refer to the Amazon DynamoDB documentation for information on setting up DynamoDB.

This code implements the idea of a generic "Session" that represents a Conversation Memory entry. For example, a "Session" could be used to represent an individual user's conversation, or a group conversation thread.

import os
import sys

import boto3

from griptape.drivers import (
    AmazonDynamoDbConversationMemoryDriver,
)
from griptape.memory.structure import ConversationMemory
from griptape.structures import Agent

if len(sys.argv) > 2:
    user_input = sys.argv[1]
    session_id = sys.argv[2]
else:
    user_input = "Hello!"  # Default input
    session_id = "session-id-123"  # Default session ID

structure = Agent(
    conversation_memory=ConversationMemory(
        conversation_memory_driver=AmazonDynamoDbConversationMemoryDriver(
            session=boto3.Session(
                aws_access_key_id=os.environ["AWS_ACCESS_KEY_ID"],
                aws_secret_access_key=os.environ["AWS_SECRET_ACCESS_KEY"],
            ),
            table_name=os.environ["DYNAMODB_TABLE_NAME"],  # The name of the DynamoDB table
            partition_key="id",  # The name of the partition key
            partition_key_value=session_id,  # The value of the partition key
            value_attribute_key="value",  # The key in the DynamoDB item that stores the memory value
        )
    )
)

print(structure.run(user_input).output_task.output.value)

Conversation Memory for an individual user:

python session.py "Hello my name is Collin." "user-id-123"
python session.py "What is my name?" "user-id-123"

> Hello Collin! How can I assist you today?
> Your name is Collin.

{
  "id": {
    "S": "user-id-123"
  },
  "value": {
    "S": "{\"type\": \"ConversationMemory\", \"runs\": [{\"type\": \"Run\", \"id\": \"8c403fb92b134b14a0af8847e52e6212\", \"input\": \"Hello my name is Collin.\", \"output\": \"Hello Collin! How can I assist you today?\"}, {\"type\": \"Run\", \"id\": \"706d9fb072ca49e192bfed7fc1964925\", \"input\": \"What is my name?\", \"output\": \"Your name is Collin.\"}], \"max_runs\": null}"
  }
}

Conversation Memory for a group of users:

python session.py "Hello my name is Zach." "group-id-122"
python session.py "And I'm Matt" "group-id-123"
python session.py "And I'm Collin, who all is here?" "group-id-123"

> Hello Zach! How can I assist you today?
> Hello Matt! Nice to meet you too. How can I help you today?
> Hello Collin! So far, we have Zach, Matt, and now you. How can I assist you all today?

{
  "id": {
    "S": "group-id-123"
  },
  "value": {
    "S": "{\"type\": \"ConversationMemory\", \"runs\": [{\"type\": \"Run\", \"id\": \"b612cdf5908845e392c026e1cf00460b\", \"input\": \"Hello my name is Zach.\", \"output\": \"Hello Zach! How can I assist you today?\"}, {\"type\": \"Run\", \"id\": \"4507988d82164cad8a288da8c984817c\", \"input\": \"And I'm Matt\", \"output\": \"Hello Matt! Nice to meet you too. How can I help you today?\"}, {\"type\": \"Run\", \"id\": \"65a70c22dae24655b312cf8eaa649bfd\", \"input\": \"And I'm Collin, who all is here?\", \"output\": \"Hello Collin! So far, we have Zach, Matt, and now you. How can I assist you all today?\"}], \"max_runs\": null}"
  }
}