operator online58 classes11 agents0 auto-submitsscope-locked
Autonomous bug bounty operator

Anyone can run ascanner.HUNTR runs the hunt.

Point it at a target you're authorized to test. HUNTR maps the attack surface, tests it class by class, validates what's real, and drafts the report. You review and submit. It never submits for you.

Validated, report-ready findingsNever auto-submitsAuthorized targets only
huntr · fire controlidle
$hunthttps://
try
ready — enter a targettested 5valid 3noise 2
CRITSQLi/items?id=✓ validated
HIGHIDOR/account/{id}✓ validated
MEDCORS/api/datadropped
HIGHSSRF/fetch?url=✓ validated
LOWXSS/search?q=dropped
coverage100%
▸report drafted — reproduction + impact · you review & submit
At a glance

Built to hunt. Measured against ground truth.

58
Vulnerability classes

Hunted end-to-end, recon to validated report.

11
Specialist agents

Recon, exploit, chaining, reporting — each a focused agent.

0
Auto-submissions

You review and send. Every finding, every time.

Why HUNTR

Scanners give noise.
Hunters give payouts.

Coverage

End-to-end hunt

Maps the surface, tests it, and keeps going until coverage is real — not a one-shot scan.

Signal

Validated, not noisy

False positives are dropped before they reach you. You read a short list of real bugs.

Output

Report-ready

Reproduction and impact come drafted with each finding. Minutes to review, not hours to write.

Control

You hold the trigger

Nothing is auto-submitted. HUNTR only runs on targets you're authorized to test.

surface map · live
drag to orbit · hover a node
Live surface map

The whole surface, one living graph.

Every host, endpoint, and parameter is a node; every edge is a path HUNTR can chain. It's how a low-severity bug turns into a critical one — the engine reasons across the whole surface instead of testing endpoints in isolation, so it finds the routes a scanner walks right past.

criticalhighvalidateddrag to orbit
Comparison

HUNTR vs. a scanner.

Same target, two very different outputs. A scanner flags; HUNTR hunts, validates, and writes it up.

HUNTR

Operator
  • Tests behavior, then validates every finding
  • Drops false positives before they reach you
  • Reproduction and impact drafted per finding
  • Covers the surface to depth — not one-shot
  • Never auto-submits — you review and send
Start hunting

Traditional scanner

Noise
  • ×Stops at surface signals (200 OK, reflected param)
  • ×Dumps everything — you sift the false positives
  • ×No reproduction or impact — you write the report
  • ×One-shot, shallow pass over the surface
  • ×Flags without judgment — confirms nothing
The status quo
FAQ

Questions, answered.

What is HUNTR?

HUNTR is an autonomous AI operator for bug bounty. You point it at a target you're authorized to test and it maps the attack surface, tests it, validates what it finds, and drafts the report. You review and submit.

Does HUNTR submit reports automatically?

No. HUNTR never auto-submits. Every finding is a draft for you to review and submit. The judgment stays human.

How is it different from a vulnerability scanner?

Scanners stop at surface signals and produce noise. HUNTR runs a real methodology end-to-end, validates each finding to drop false positives, and delivers report-ready vulnerabilities with reproduction and impact.

Does it only run against authorized targets?

Yes. HUNTR only runs against targets you're authorized to test, and it respects program scope. It's a tool for authorized security research.

Stop scanning. Start hunting.

Create your account, load credits, and point HUNTR at a target you're authorized to test. We run the hunt for you.