Hunt faster than the attacker.

Find who's already in.Stop who's next.

Huntbase is an autonomous threat hunter. When a new threat is published, Scout checks months of history across the tools you already run and your endpoints, responds within the rules you set, and proves it's gone.

Where we are now

The clock starts when they get in.

MTTR is usually measured from the alert. For a quiet intrusion, the clock started weeks earlier, and nothing fired to say so.

If your team only looks when an alarm rings, nobody looks at all.

“Boards should have confidence that their organisation can detect and investigate emerging cyber threats, including activity that existing security alerts may miss.”

National Cyber Security Centre (NCSC), Why threat hunting should matter to every board, November 2026 (opens in a new tab)

Where you want to be

You know who got in, you’ve stopped them, and you can prove it.

  • Every new threat is checked against your past the day it’s published.
  • If someone got in, you know where, when and how, and the response is already under way within your rules.
  • You can prove the threat is gone, and the same attack can’t land twice.
  • Your team sets the rules and reviews the outcomes. Scout writes the queries at midnight.

How Huntbase gets you there

How a hunt runs, from new intel to proof

Scout hunts back through your history, acts within the rules you set, and leaves evidence you can check at every step.

  1. Look back

    Find who's already in.

    New advisory, IOC or CVE? Scout checks it against weeks to months of history in your SIEM, EDR, cloud and data lakes, and on each host through the endpoint agent. The answer comes with evidence: hosts, users and a timeline.

    One query answered by four connected tools, with the events over time and the results in a single list
    One question asked of four connected tools at once: 142 sign-in events, over time, in one list.

A patch closes the door. It doesn’t tell you who already walked through it.

  1. Respond

    Shut the door, proportionately.

    Scout acts within your response rules: kill the process, quarantine the file, isolate the host only if the threat is still live. Anything outside your rules waits for a person, two by default. Every action can be undone.

    Response · WKS-2231

    Killed process · quarantined file · WKS-2231

    • Killed rundll32.exePID 4812
    • Quarantined cache.dllC:\Users\Public
    • Isolate host: not needed, threat no longer live
    Rule: contain known-bad on workstationsUndo
  2. Verify

    Prove it's gone.

    Scout re-hunts to confirm the host is clean and records who acted, on whose behalf, and under which rule.

    Verify · WKS-2231

    Re-hunt clean

    0 hits since containment · 3 sources checked

    ActorScout on behalf of m.reyes
    Rulecontain known-bad on workstations
    Recorded14 Nov · 09:42 UTC
  3. Keep watching

    Stop who's next.

    The hunt becomes a watcher, backtested against 1,250+ attack recordings and run in shadow before it goes live. Scout also points out where you're exposed, so you can patch first.

    Watcher · cache.dll side-load

    rundll32 loads an unsigned DLL from a public folder

    1. Backtested
    2. Shadow
    3. On

    Backtested on 1,250+ attack recordings · 7 days in shadow

    Patch first: 7-Zip 24.07 on 3 hostsCVE-2025-0411 · fixed in 24.09

What you want, and what you need

You want the answer. You’re accountable for the evidence.

Huntbase gives you both: the answer the day a threat is published, and the record behind it.

WhoWhat they wantWhat they need
SOC and hunt leadsFor SOC teamsA calm answer when the CISO forwards the advisory. Fewer consoles. No lost weekend.Evidence behind every answer, and every new threat in the news checked against their own history.
CISOsTrust centerTo say “we checked” and mean it.Proof for the board, auditors and regulators: what was checked, what was found, and which person or rule allowed each action.
MSSPs and IR firmsFor MSSPsTo serve more clients with the same team.A defensible record of every hunt and action, kept separately for each client.

Alongside the tools you run

They triage alerts.
Huntbase hunts.

Huntbase answers the questions nothing fired for. Keep the tools you trust: Huntbase reads their data and hands the evidence back as a hunt report, an export or over the API.

  • Your SIEM

    Holds what was logged.
  • Your EDR

    Works what fired on the endpoint.
  • Your AI SOC

    Triages the alerts that fire.
  • Huntbase

    Hunts for what already got in, and stops it coming back.

A few of the 158 products Huntbase integrates with.

  • Splunk
  • Microsoft Sentinel
  • CrowdStrike Falcon
  • Elastic Security
  • SentinelOne
  • Okta
  • Amazon AWS
  • Google Cloud Platform

True today

Open, in use, and built by people who have worked the breach.

  • documented integrations

  • open hunt playbooks on the Huntbase Hub

  • public attack recordings that watchers are tested against

  • autonomy levels (Guide, Collaborator, Operator), with every action audited and undoable

Your hunts stay portable: the hunt.md format is an open spec (opens in a new tab), and Huntbase speaks Sigma, OCSF, STIX/TAXII and CACAO. Browse the Hub (opens in a new tab). Scout doesn’t train shared models on your data.

Already in?
Find out today.

Sign up and watch Scout hunt a new threat through the sample workspace, or book a demo and bring the threat you need checked.

  1. Sign up.

  2. Run a guided hunt on the sample workspace. No connector needed.

  3. Connect your own sources when you're ready.