Skip to content

Add demo agents for TLCS learner memory demonstration#1

Open
dwright-snhu wants to merge 1 commit intomainfrom
demo-agents-local
Open

Add demo agents for TLCS learner memory demonstration#1
dwright-snhu wants to merge 1 commit intomainfrom
demo-agents-local

Conversation

@dwright-snhu
Copy link
Collaborator

Summary

  • Adds LangGraph-based AI tutors (Math Tutor, Writing Coach) that demonstrate TLCS integration
  • Demo runs locally and connects to deployed TLCS infrastructure (no ECS deployment needed)
  • Shows real AI conversations with automatic memory extraction and Caliper format

What's included

  • demo/agents/ - Self-contained Python demo with LangGraph agents
  • demo_alice_conversation.py - Main demo showing full flow (Alice story from Pre-read Section 3)
  • Memory extraction, Caliper export, scope-based filtering between agents
  • Verbose output on by default for presentation evidence

Running the demo

cd demo/agents
uv sync
export OPENAI_API_KEY=your-key
uv run python scripts/demo_alice_conversation.py --mock

For deployed TLCS: use --real flag with TLCS environment variables.

Test plan

  • Demo runs with --mock flag (simulated TLCS)
  • Verbose output shows Caliper JSON, filtered memories, injected context
  • --quiet flag suppresses verbose output
  • Demo runs with --real flag against deployed TLCS

🤖 Generated with Claude Code

LangGraph-based AI tutors (Math Tutor, Writing Coach) that demonstrate
TLCS integration:

- Real AI conversations with automatic memory extraction
- Caliper event format for pushing memories to TLCS
- Cross-tool memory sharing with scope-based filtering
- Alice story demo showing Pre-read Section 3 scenario

Demo runs locally with --mock (simulated TLCS) or --real (deployed TLCS).
Verbose output is on by default for presentations.

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
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.

2 participants