TP / Open source

Generative AI / Hackathon prototype

CORTEGE

Explore household-aware fraud assessment with companion agents.

View source ↗Try it ↓
  1. 01Event + evidence
  2. 02Household agents
  3. 03Explainable risk case

When to use it

Demonstrate a household fraud case using a simulated event or phone call plus a message excerpt, suspicious URL, or screenshot note.

Get started

Requirements

Node.js 18+, npm, and an Anthropic API key. Twilio integration requires the separate deployment setup.

Terminal
git clone https://github.com/taylorparsons/cortege-hackathon.git
cd cortege-hackathon
./run-local.sh
# Add ANTHROPIC_API_KEY when prompted

How to use

  1. Open the frontend at http://localhost:5173. The backend runs on port 3001.
  2. Choose or create a household and select its members.
  3. Use the event simulator or Live Feed to add a test event and a manual evidence item.
  4. Review the generated case severity, signals, rationale, and recommended action. Stop the app with ./stop-local.sh.

Under the hood

The React frontend connects to an Express backend through REST and WebSocket. An orchestrator manages member-specific agents and Claude assessments. This is a prototype with heuristics, not verified fraud prevention or image/video authenticity detection.

Open source recognition

Meet the maintainers and contributors, view dependency sources, and read original license notices ↗

Source documentation

Setup instructions are taken from the public repository. Dependencies, API availability, and scripts can change; check the source README for updates.

Questions or issues ↗

Explore another workflow

Browse by use case ↗