Skip to content

[Use Case] Memory-Powered Customer Support Agent with Short-Term + Long-Term AgentCore Memory #1735

Description

@Ashutosh0x

Use Case Proposal

I'd like to contribute a Memory-Powered Customer Support Agent use case demonstrating AgentCore Memory in an end-to-end conversational agent.

What it demonstrates:

  • Short-term memory (session context via events)
  • Long-term memory (semantic extraction with CustomerFacts and IssueHistory strategies)
  • AgentCore Memory SDK (MemoryClient) with Strands Agents SDK
  • Cross-session customer recognition and personalization
  • Memory-backed tools for customer history recall, ticket creation, and order lookup

Technical stack:

  • Python + Strands Agents SDK
  • AgentCore Memory (create_memory_and_wait, create_event, retrieve_memories)
  • Semantic strategies: CustomerFacts, IssueHistory
  • Unit tests with pytest

Location:

02-use-cases/01-conversational-agents/memory-powered-customer-support/

PR:

#1734

I've followed the contributing guidelines and the 02-use-cases template format.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions