Skip to content

Simulates real-world attacks (SSH brute-force, Mimikatz) and detects them using a custom Python tool with Shodan API and an ELK Stack SIEM pipeline. Includes log parsing, dashboards, and detection rules for end-to-end threat simulation and response.

Notifications You must be signed in to change notification settings

mohakrudrakshh/-End-to-End-Threat-Simulation-and-SIEM-Detection-Pipeline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Threat Simulation & SIEM Detection Pipeline (TS-SDP)

image

Threat Simulation & SIEM Detection Pipeline (TS-SDP) is an end-to-end, modular security lab that simulates real-world attacks and detects them using a fully integrated ELK Stack. It enables hands-on experience with threat simulation, detection engineering, log analysis, and SOC workflows.

Lab Diagram

image

Lab Infrastructure Diagram

image

The diagram above outlines the architecture of TS-SDP, which includes the following key components:

  • Sources: Sysmon, Suricata, Filebeat, and simulated attack logs from Linux/Windows endpoints.
  • Monitoring and Aggregation: Filebeat ships logs to Logstash for parsing and forwarding to Elasticsearch.
  • Output Tools: Kibana dashboards visualize brute-force attempts, process anomalies, and credential misuse.

Key Elements:

  1. Sources:

    • Sysmon
    • Suricata
    • Filebeat (Linux/Windows)
    • Simulated logs via Python script
  2. Monitoring & Data Ingestion:

    • Logstash (Grok Filters)
    • Elasticsearch
  3. Output & Analysis:

    • Kibana Dashboards
    • Alert rules and visual queries

Each layer of the pipeline is designed to teach you how real-world telemetry is generated, parsed, and acted upon in a SOC environment.

Capabilities

  • Realistic Attack Simulation: Includes SSH brute-force and credential misuse using a custom Python + Shodan tool.
  • Custom Log Ingestion: Parses logs from simulated attacks using Grok in Logstash.
  • Interactive Dashboards: Kibana visualizes live attack patterns and detection metrics.
  • Rule Tuning Practice: Simulated alerts allow hands-on experience in minimizing false positives.
  • Modular & Scalable: Easily extendable to other attacks or log sources.

System Architecture

Components

  • SSH Brute-force Tool: Python script using Shodan API and Paramiko for parallelized scanning.
  • Linux/Windows Clients: Hosts that generate logs using Filebeat, Sysmon, Suricata.
  • Log Aggregator: Logstash configures parsing and routes to Elasticsearch.
  • SIEM Frontend: Kibana for dashboards and alerting.

Key Takeaways

  • Building end-to-end threat detection workflows.
  • Writing and tuning Grok patterns in Logstash.
  • Understanding attack telemetry in real-time.
  • Developing detection logic and visual queries.
  • Practicing alert triage based on live data.
  • Hands-on SIEM engineering and blue team skills.

Tooling

  • Python: For simulating brute-force attacks.
  • Shodan API: Real-time host discovery for scanning.
  • Paramiko: SSH brute-force logic.
  • Sysmon & Suricata: Endpoint and network telemetry.
  • Filebeat: Log shipper for Linux/Windows.
  • Logstash: Parsing and forwarding pipeline.
  • Elasticsearch: Storage and search engine.
  • Kibana: Dashboard and query visualization.

System Requirements

  • Python 3.8+ (for attack tools)
  • ELK Stack (7.x+) running locally or on VM
  • Filebeat + Sysmon + Suricata on monitored machines
  • Shodan API Key (for scanning public targets)
  • Ubuntu/Windows VMs for simulation

Integrated Solutions

  • Kibana: Visualize brute-force, credential misuse, and anomaly detection.
  • Logstash: Parse logs with Grok, enrich and forward to Elasticsearch.
  • Filebeat: Stream logs from endpoints to Logstash.
  • Custom Attack Tool: Runs SSH brute-force + Shodan scanning.
  • Sysmon & Suricata: Capture endpoint and network-level events.

Configuration Options

TS-SDP is modular and adaptable for:

  • Local VM-based testing (ELK stack running on same host)
  • Cloud-based log collection (via Filebeat + public IPs)
  • Tool replacement or extension (e.g., add Caldera, Atomic Red Team)
  • Custom rules for detections based on simulated attacks

Usage Scenarios

  • SOC Lab Simulation: Emulate a real detection pipeline.
  • Threat Detection Testing: Validate your SIEM and alerting setup.
  • Detection Engineering Practice: Learn to write Grok filters and tune rules.
  • Cybersecurity Education: Train students on real telemetry and alerts.
  • Portfolio Development: Add a practical, hands-on project to your GitHub.

Credits

This project was built independently by Rudraksh Gupta to demonstrate hands-on threat simulation, detection, and response workflows. Inspired by community contributions and the goal of enabling better blue team readiness.

About

Simulates real-world attacks (SSH brute-force, Mimikatz) and detects them using a custom Python tool with Shodan API and an ELK Stack SIEM pipeline. Includes log parsing, dashboards, and detection rules for end-to-end threat simulation and response.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages