What happens when an AI consultancy becomes the target?

Neural is an AI & Active Directory hybrid lab that puts you inside a corporate environment where AI-driven workflows are deeply integrated with the underlying infrastructure.

That creates attack paths that cross the boundary between AI and traditional enterprise security. Your goal is to find them.

Compromise AI integrations, abuse Active Directory, move across trust boundaries, chain weaknesses together and work your way through the environment.

Neural puts the AI attack surface where it belongs: in the middle of a real enterprise attack path, alongside the infrastructure security teams already know they have to defend.

Can you uncover the weaknesses hiding behind Neural’s AI-powered infrastructure?

It’s not enough to know how to use an AI tool. You need to know what to delegate, what to verify, and when to intervene, especially when an AI agent is moving quickly, operating across multiple tools, or confidently getting something wrong.

Our new learning paths are built around that shift.

Build the mental models you need to understand what AI systems are actually doing and where they can go wrong.

Learn how models, retrieval, agents, memory, context, and tool calls work, and how to verify AI output against ground truth.

Run offensive workflows with agents while keeping the engagement under your control.

Learn to direct agents across reconnaissance, exploitation, and post-exploitation while maintaining scope, verifying their proposals, protecting OPSEC, and knowing when to intervene.

Use AI agents to accelerate investigation and response without letting speed replace analyst judgment.

Learn to evaluate agent conclusions against evidence, catch confident-but-wrong analysis, calibrate autonomy, and recover when an agent takes the wrong path.

The first modules are only the beginning. More content will build on these foundations as the paths expand.

Start building the skills to work with AI as part of the team, not just as a tool.

Effective adversary emulation begins long before the first packet is sent.

This module covers the planning and reconnaissance that precede a red team engagement, exploring how operators decide which adversary to emulate and how they gather information about a target while remaining undetected. As organizations mature their defenses, the ability to plan a realistic engagement and conduct reconnaissance without being detected is what separates a convincing emulation from a noisy one.

Train across the entire cyber operational spectrum with a single membership. HTB PRO combines enterprise-grade offensive and defensive hands-on training directly inside HTB Labs, giving you full VIP+ access, 10 flagship Pro Lab environments, and upcoming SOC Range simulations.

What is included:

HTB PRO is available now on HTB Labs for $49 per month or $490 per year. Existing VIP+ options remain active alongside this subscription.

Learn more here.

👉 Get started with HTB PRO here

The 1st HTB Enterprise Season comes to an end with Uplink, a Hard Windows Machine.

Investigate a compromised satellite ground segment, uncover weaknesses in its command-signing pipeline, and determine whether forged commands can reach privileged execution paths.

Skills covered:

New to Operation Red Horizon? Catch up on previous scenarios

In this special Season, you will be running a red team engagement, set at HTB International Airport, and testing various assets across an airport network that might be vulnerable.

9 weeks. 9 Machines. One leaderboard. Kicks off on September 26th!

HTB Seasons brings competitive hacking to the forefront. Complete weekly cybersecurity labs, earn points, and climb the Seasonal leaderboard as the challenges roll out.

It’s free to join and open to the entire HTB community.

The final Sherlock of the season is live on HTB Enterprise Platform.

A forged beacon order has been staged inside a satellite broadcast queue using CFDP. You must analyse the uplink PCAP, spot the fake transmission, and stop an adversary from hijacking the fleet before the window closes.

The final Sherlock of the season is waiting...

We are bringing 150+ investigation scenarios from the LetsDefend platform directly to HTB Labs. We are releasing them as 16 new retired Sherlocks every week from the end of August 2026 until October 15th. Each batch of Sherlocks will be free to play for a full month following its release date.

This gives blue teamers and defenders a steady stream of fresh, real-world analytical scenarios to investigate and solve.

Every CTF you play now contributes XP toward your HTB progression, reflected currently on your HTB Profile. Since CTFs are team efforts, XP is split among the participating team members.

Streak Protocol is a new CTF mini-game built for quick, competitive knowledge testing.

Answer cybersecurity questions against the clock, build your streak, and compete with other players on the event leaderboard. The game is powered by 1,300+ questions generated from Marketplace content and is available on both desktop and mobile.

We've just refreshed the HTB Profile experience to give you a cleaner, more streamlined way to showcase your cybersecurity achievements.