Cyber Ranges
Cyber ranges are deliberately vulnerable lab targets you can point a Cracken operation at to see how the platform works before you run it against your own systems. Each range is a self-contained environment — a web app, an AI-backed app, or a small network — with planted weaknesses for an operation to discover, exploit, and chain.
Every target listed here is a disposable, isolated lab environment provided for testing Cracken. You are authorized to assess these specific targets. Do not point Cracken at any system you do not own or have explicit written permission to test.
How a range run works
Every range uses the same workflow, so once you have run one you can run any of them:
- Create a realm to scope the run, or open an existing one.
- Connect a Tentacle with network reach to the target. The Tentacle is where the operation's tool actions execute, so it must be able to reach the range's IP and port.
- Create an operation in that realm. Choose a starting point:
- From a Playbook — pick the built-in methodology that matches the range (for example Web App Pentest for the web ranges, Network Pentest for the infrastructure range).
- From scratch — describe the goal and let the agent plan the approach.
- Set the target to the range's address (each range page lists it).
- Run it. The agent reasons, proposes actions, and — depending on the operation's autonomy mode — either asks for approval or executes on its own.
As the operation runs, confirmed findings and the relationships between hosts, services, and credentials accumulate in the Cybergraph, captured tool output is saved as artifacts, and the agent writes a report you can download.
Available ranges
| Range | Target | Best Playbook | What it exercises |
|---|---|---|---|
| Web app range | https://77.42.79.6/ | Web App Pentest | Multi-vector web testing: injection, XSS, CSRF, IDOR, RCE, auth weaknesses |
| E-commerce web range | http://44.223.238.255/ | Web App Pentest | Black-box assessment of a storefront with reflected input and numeric object references |
| AI assistant range | http://107.23.14.14:8000/login | Web App Pentest | An LLM-backed app behind a login — web auth plus AI-application weaknesses |
| Monolith terminal range | http://23.88.113.85/ | Web App Pentest | Recon and hidden-endpoint discovery against a locked-down terminal app |
| Full estate compromise range | ssh kali@32.192.161.20 | Network Pentest | A multi-host network: entry through a web app, then escalation across the estate |
| Identity provider range | http://idp-cyberrange-dev-31786426.eu-central-1.elb.amazonaws.com/ | Web App Pentest | An SSO portal over Keycloak: JWT validation, BOLA/IDOR, tenant isolation, OAuth redirect, and SCIM provisioning weaknesses |
The web ranges run on plain HTTP or self-signed HTTPS, so expect a certificate warning on the
https:// target — that is the lab, not a problem with Cracken.
Next steps
- Quick Start — set up Cracken and connect your first Tentacle.
- Operations — how an operation runs, and how to approve or deny actions.
- Playbooks — the methodologies that drive a range run.
- Web app range — start with the multi-vector web target.