Full Hunt

One Click, a Whole Hunt

Select a program and click Start full hunt. The AI gets the whole program: its scope, rules, assets and what is new since the last recon. Then it works through a complete hunt step by step, posting short progress notes in the chat as it goes.

For the bigger jobs it brings help. It uses CoHunters and sub-agents to work in parallel, so recon, testing and verification happen at the same time.

Stays Inside Your Scope

The AI reads the program's scope, out-of-scope list and rules of engagement before it starts, and works inside them. Keep chatting to steer it, for example "focus on the API first", or press Stop at any time.

Hunt What Changed

First-to-find bugs usually sit in what is new. After a recon run, click Hunt the new assets and the list of new and changed assets goes straight to the AI to investigate first.

Start From Nothing

No program yet? Open an empty chat, type a target such as example.com into the Start a hunt card, and XHack adds it to Programs and starts mapping its attack surface straight away.

Asks When It Matters, or Not at All

When the AI needs something only you know, such as test credentials or an unclear scope boundary, it asks with a quick question card. Not answering in 2 minutes? It carries on with its stated default. Want it to run without asking? Tell it to go full, and it decides for itself and notes its assumptions.

Findings You Can Submit

Every finding is reproduced before it is logged, so you are not chasing false positives. Each one can be written up as a report ready to submit to HackerOne or Bugcrowd. Read it through, and it is ready to send.

Learn More

• Start a full hunt
• Start a hunt from an empty chat
• Questions from the AI
• Go full mode
• Findings & Reports