Skip to content

Havoc Machine is a Synthetic Customer Chaos Lab that stress-tests customer support chatbots through AI-generated adversarial conversations. It automatically generates 1,000+ realistic chaos scenarios to expose policy vulnerabilities and refund leakage before impacting real customers.

Notifications You must be signed in to change notification settings

arpankumarde/havoc-machine

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Havoc Machine

Synthetic Customer Chaos Lab - Stress-test customer support systems through AI-generated adversarial conversations that expose policy vulnerabilities, refund leakage, and chatbot weaknesses before they impact real customers.

Problem Statement

Support operations face a critical gap between demo performance and real-world resilience. Generative AI-powered chatbots excel in controlled environments but struggle under authentic pressure scenarios including angry customers, code-mixed languages, incomplete information, and deliberate policy exploitation. Current testing methods cannot economically simulate thousands of realistic support conversations, resulting in:

  • Inconsistent policy application
  • Revenue leakage through exploited loopholes
  • Degraded customer experiences that erode brand trust

Features

  • AI-Generated Adversarial Conversations: Simulate thousands of realistic support scenarios
  • Policy Vulnerability Detection: Identify weak points in refund policies, replacement workflows, and COD return handling
  • Multi-Scenario Testing: Test against angry customers, code-mixed languages, incomplete information, and policy exploitation attempts
  • Parallel Execution: Run multiple adversarial sessions simultaneously for comprehensive testing
  • Real-time Dashboard: Monitor test runs and analyze results through web interface

Architecture

  • Server (server/): FastAPI backend with LangChain, OpenAI, and MongoDB for conversation orchestration
  • Client (client/): Next.js frontend with Prisma and PostgreSQL for dashboard and reporting
  • CLI (cli/): Python CLI tool for running parallel adversarial red teaming sessions

Quick Start

Prerequisites

  • Python 3.12+ (for server and CLI)
  • Node.js 18+ (for client)
  • PostgreSQL database
  • MongoDB database
  • OpenAI API key

Configuration

Create .env files in respective directories with required API keys and database connection strings.

Tech Stack

  • Backend: FastAPI, LangChain, OpenAI, MongoDB, PostgreSQL
  • Frontend: Next.js 16, React 19, Prisma, Tailwind CSS
  • CLI: Python with Rich for terminal UI

About

Havoc Machine is a Synthetic Customer Chaos Lab that stress-tests customer support chatbots through AI-generated adversarial conversations. It automatically generates 1,000+ realistic chaos scenarios to expose policy vulnerabilities and refund leakage before impacting real customers.

Resources

Stars

Watchers

Forks

Contributors 3

  •  
  •  
  •