Xseth

Your AI recon partner for offensive security

6 followers

Your AI recon partner for offensive security

6 followers

Xseth maps your external attack surface like an attacker would then triages what matters. Autonomous recon across subdomains, services, and exposure at machine speed, with every attack path narrated so you see how findings chain. A force-multiplier for security teams and operators buried in scanner noise. Stop chasing false positives. Find what attackers find, first.

Free Options

Launch tags:Artificial Intelligence•Tech

Launch Team

CotypistLocal AI Autocomplete in your voice, anywhere on your Mac

Promoted

Maker

📌

I'm BENKRIEF Syef , co-founder and CEO of Xseth. This started from a frustration anyone in offensive security knows intimately: the hard part was never finding things. It's drowning in scanner output and burning hours separating real exposure from noise. Most tools hand you a 400-line report and wish you luck and the signal that actually matters gets buried. So the problem we set out to solve was simple to state: get an operator from "here's everything" to "here's what matters, and here's how it chains into a real attack" without the manual slog in between. Honestly, our thinking evolved a lot along the way. We started out describing Xseth as an autonomous pentesting platform and we were quietly overclaiming. The more we built, the more we realized the genuinely valuable, defensible thing was being the best recon-and-triage engine in the world: mapping your external attack surface across subdomains, services, and exposure at machine speed, then narrating how findings chain into an attack path. Force-multiplier, not magic box. That honesty with ourselves is what made the product real. The next piece the part we're building right now is a deterministic validation layer that confirms a finding is exploitable rather than just flagging it. That's the frontier, and it's exactly why we want beta users in the loop early. So we're opening a small beta and looking for a handful of design partners: bug bounty hunters, solo operators, red teamers people who live in this problem and will tell us where it breaks. Free access, direct line to the founders. My co-founder and CTO, Taleb, will happily go deep on the architecture with anyone curious. If that's you, drop a comment or DM me. Ask us anything.

Report

5mo ago

Forum Threads

p/xseth

•

3d ago

Help us design a security report you'd actually trust (free scan inside)

Most security reports fail the same way: 40 pages of "possible" issues, and nobody can tell which three are real. We built Xseth to fix that, and now we want your help deciding what the report itself should look like.

Quick recap of what Xseth does: it looks at your web app from the outside, the way an attacker would, narrows thousands of URLs down to the few worth testing, then validates each one. No install, no code access, no accounts. A finding is only marked PROVEN when a deterministic check confirms it, never on an AI's word alone. Everything else is labelled refuted, inconclusive, not tested or false positive.

p/xseth

•

1mo ago

6 candidates, 1 proven: we're opening the Xseth beta

We launched Xseth here a while back and have shipped a lot since, so here's what it actually does now, in one GIF. It runs external recon on a domain, cuts thousands of URLs down to the handful worth testing, then checks each one. A finding only gets marked PROVEN when a deterministic check confirms it, and everything else stays honestly labeled instead of padding a report. The GIF is an illustrative run on a demo target. We're opening the beta to bug bounty hunters and to teams shipping fast, including AI-built apps on Lovable, Bolt or v0. It's external only: no accounts, no code access, nothing to install. All we need is a domain you own or are authorized to test. In return, we'd love a few minutes of your time or one honest email on what was right, what was missing and what you'd never use. Drop a comment or reach us at xseth.com.

View all