Skip to content

Conversation

omrsamer
Copy link
Contributor

@omrsamer omrsamer commented Oct 8, 2025

  • Added setup_memory.py script for one-time memory creation
  • Implemented session-based S3 file persistence
  • Added memory viewing scripts (conversation history and long-term memories)
  • Updated memory_manager.py to load from SSM instead of creating
  • Enhanced agent.py with proper memory context loading
  • Added comprehensive MEMORY_SETUP.md documentation
  • Fixed script parameter passing for actor_id and session_id

Fixes #

- Fixed multiple memory ID creation issue by separating setup from runtime
- Added setup_memory.py script for one-time memory creation
- Implemented session-based S3 file persistence
- Added memory viewing scripts (conversation history and long-term memories)
- Updated memory_manager.py to load from SSM instead of creating
- Enhanced agent.py with proper memory context loading
- Added comprehensive MEMORY_SETUP.md documentation
- Fixed script parameter passing for actor_id and session_id
- Integrated memory setup into deploy.sh (idempotent, SSM-first approach)
- Fixed multiple memory IDs issue by using SSM Parameter Store as source of truth
- Added session summaries support using list_sessions() and list_memory_records()
- Updated Streamlit UI Tab 3 to display conversation summaries
- Updated view_memories.sh CLI to show session summaries
- Added get_session_summaries() method to memory_manager.py
- Fixed turn numbering in Streamlit (most recent turn shows highest number)
- Removed setup_memory.py (functionality moved to deploy.sh)
- Added comprehensive README.md with updated documentation
- Added .gitignore for generated files
- Follows customer-support-assistant pattern from AWS samples
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant