Connic
Back to templates

Kafka Fraud Detector

Score transactions for fraud in real-time on Kafka streams. Uses velocity analysis, geo-anomaly detection, and pattern matching per customer.

securityadvancedKafkaView on GitHub
connic init my-project --templates=kafka-fraud-detector
Quickstart

Overview

Consumes transactions from a Kafka topic and scores each for fraud risk using velocity analysis, geo-anomaly detection, and knowledge base pattern matching. Conditional tools restrict admin_override access based on context. Concurrency control queues transactions by customer_id to prevent race conditions. High-risk alerts are formatted by a tool agent and published to a Kafka outbound topic.

Use cases

Payment fraud detection

Score every transaction in real-time for fraud signals like impossible travel, velocity spikes, and known patterns.

Risk-based routing

Route transactions to different review queues based on risk level: auto-approve, flag, hold, or block.

Pattern learning

Store novel fraud patterns in the knowledge base so detection improves over time.

Architecture

Kafka (transactions)
fraud-scorer
Knowledge Base
fraud-escalator
Kafka (alerts)

Scaffolded project structure

Running connic init my-project --templates=kafka-fraud-detector creates this file tree.

kafka-fraud-detector/
  agents/
    fraud-scorer.yaml
    fraud-escalator.yaml
  tools/
    fraud_tools.py
  middleware/
    fraud-scorer.py
  schemas/
    fraud-assessment.json
  requirements.txt
  README.md

Get started

Install the template, create a Connic project, and deploy. Choose Git (automatic on push) or CLI (works with any provider).

Prerequisites

  • Python 3.10+
  • A Connic account (create a project first)
  • API key for your LLM provider (e.g. Gemini, OpenAI) to add in project variables
Create project
1

Install and scaffold

Install the SDK and create a project from this template.

terminal
pip install connic-composer-sdk
connic init my-project --templates=kafka-fraud-detector

Then cd my-project

2

Deploy

Pick your deployment method. Git auto-deploys on push; CLI works with GitLab, Bitbucket, or no Git.

Git integration

  1. In Connic: Project Settings → Git Repository, connect your GitHub repo
  2. Settings → Environments: map branch (e.g. main) to Production
  3. Push your scaffolded project to that repo
terminal
git add .
git commit -m "Add Kafka Fraud Detector template"
git push origin main

CLI deploy

  1. In Connic: Project Settings → CLI, create an API key and copy project ID
  2. Run connic login in your project folder
  3. connic test to try with hot-reload, or connic deploy for production
terminal
connic login
connic test    # Ephemeral dev env with hot-reload
connic deploy # Deploy to production
3

Connect and configure

Add a Kafka inbound connector for the transactions topic and a Kafka outbound connector for fraud alerts. Configure bootstrap servers and optional SASL authentication. Add your LLM provider API key in Project Settings → LLM Provider API Keys.

Full connector setup guideQuickstart guideConnector docs

Template source

Browse the full template, contribute improvements, or fork for your own use.

connic-org/connic-awesome-agents/tree/main/kafka-fraud-detector

Explore other templates

View all

Research Assistant

Break complex research questions into sub-tasks, dispatch specialist agents in parallel, and synthesize web and internal data into structured, scored reports.

SQS Order Processor

Validate and fulfill orders from SQS queues with fraud detection, inventory checks, and concurrency keys. Results publish to an outbound queue.

Telegram Personal Assistant

Multimodal Telegram assistant that processes text, images, voice, video, and documents. Searches the web, saves notes and knowledge, and schedules follow-up reminders with persistent sessions.

Lead Enricher

Enrich and qualify new signups by researching their company and scoring fit against your ideal customer profile. Saves qualified leads.