Skip to main content

Conclude a hunt

A hunt ends with a verdict: what you found, and why you think so. You record it in one place, the Conclude sheet, which also makes you settle what the hunt left open, pick the evidence behind the verdict, choose the write-up and decide the follow-ups. On this page you see how the checks' verdict is worked out, conclude a hunt, read the verdict cell, reopen or archive a hunt, and save, export and reuse it.

Two verdicts: the checks' and yours​

A hunt can carry two verdicts, and Huntbase never lets one overwrite the other.

  • The checks' verdict is worked out from the searches the hunt (and its Scout chats) actually ran. Scout can't set it or argue it into something else. It shows in the answer box at the top of the hunt, in the Brief and under Scout's answers. See How the checks' verdict is decided.
  • Your verdict is the one you record when you conclude. It's the outcome of record.

Once you've recorded a verdict, it leads everywhere as Analyst: Confirmed (or whichever you picked). The checks' verdict stays underneath in muted text, for example From the checks: Can't confirm — some checks found something, but not enough to confirm it., whether or not the two agree. Before you record one, the checks' verdict is shown on its own.

How the checks' verdict is decided​

Checks' verdictWhen it applies
AffectedA check found matching activity in a live or real source, and it isn't only old data.
Can't confirmSomething was found, but only in old data or sample data, while the live tools that searched found nothing.
Not affectedEvery indicator in the question was searched, every source answered, the period was long enough, and nothing matched.
No evidence in what we can seeEverything that ran came back clean, but something wasn't searched: a source failed, was held for approval or has no data for the period, or the period was shorter than asked. That isn't proof of absence.
Can't tell yetNothing has run yet, or every check failed.

The checks' verdict only changes when a new check arrives. Asking Scout again, or Scout writing something different, doesn't change it. When it does change, it says why, for example Changed because a newer DNS check found evil.example.

Verdicts​

Every place that records or shows a hunt's outcome uses the same five verdicts, in this order:

VerdictMeaning
ConfirmedReal malicious activity. Needs at least one cell with results as its basis.
BenignReal activity, but expected or authorised.
False positiveThe detection was wrong.
ClearWe checked and it isn't there.
InconclusiveCouldn't settle it; the gap is recorded.

Some verdicts can't be picked on every hunt. A verdict you can't use stays in the list, dimmed, and says why when you pick it:

  • Clear needs at least one check that ran without an error; a check that ran and returned 0 rows counts. Until then it says why, for example None of this hunt's 7 checks has run yet or All 3 checks failed. Checks that failed or never ran are open issues: run them or acknowledge them, then Clear. Clear is blocked outright when the results hold risk signals.
  • Confirmed needs evidence: at least one evidence cell that returned rows. Cells that returned 0 rows can't carry a Confirmed verdict on their own.
  • For a CVE, vulnerable-version or exposure finding, the hints read in those terms, for example Confirmed — the exposure is real or Not affected — the version or host doesn't match.
  • False positive over risk signals asks Why don't the risk signals count? and won't record without an answer.

Open the Conclude sheet​

Open it from any of these:

  • Conclude in the hunt header (Edit conclusion once the hunt is concluded);
  • Conclude in the verdict cell at the end of the notebook;
  • Conclude as Clear… in the details pane's Brief, which opens the sheet with Clear picked;
  • Conclude… under a Scout answer that states a verdict, in a chat attached to the hunt. It suggests that verdict and fills in the start of the answer as your reason;
  • an item in the Inbox.

The sheet opens on the right with five numbered sections. Press 1 to 5 to jump between them, and ⌘↵ or Ctrl+↵ to conclude.

1. Verdict​

Pick one of the five verdicts and say why in a line or two (Why? One or two lines your team (or the customer) can read — required). Your reason becomes the hunt's headline.

If Scout drafted a verdict, the section says so (Scout drafted Benign, citing 2 cells. You decide what is recorded.) with Start from Scout's draft. If a cell ran again after Scout wrote the draft, it warns you to check the draft against the new results. When the results hold risk signals, they're listed here; click one to go to its cell. A response action that failed, such as an isolation, is called out before you record.

2. Open issues​

Open issues lists what the hunt left unsettled: cells that failed, didn't run, are still running or are out of date, checkpoints nobody answered, and failed actions. For each one, click Resolve to go to the cell and deal with it, or tick Acknowledge to record it as a known gap. Acknowledge all ticks them all. You can't conclude until every issue is resolved or acknowledged. Acknowledged issues are recorded with the verdict and listed as limitations in the customer report.

3. Evidence​

Tick the cells the verdict rests on. Cells that returned rows are ticked to start with, and the section counts the tagged rows. If no cell with results is included, the sheet warns that the evidence is thin and the report will say the outcome is provisional.

4. Write-up​

Pick the Audience:

  • Internal: your reason is the headline everyone sees. Add Notes for the team (never in the customer copy), such as a ticket number or a caveat; they are saved with the finding. The internal report adds the method, the queries and Scout's reasoning: open it from here (Open internal report, or Go to the report tab when there isn't one yet). See Hunt reports.
  • Customer copy asks for What happened (for the customer) and What they need to do (one per line), with Preview customer report. See Customer report.

5. Follow-ups​

Each follow-up has a default for the verdict you picked. Chosen follow-ups are recorded with the verdict; you start each one yourself after concluding.

Follow-upWhat it offers
Contain with the hosts named (or Contain the affected systems)After concluding, a link to contain each host from its endpoint page, or a note on why it can't be done here. A host on the sample fleet reads Open with the host's name and Sample fleet: response actions are disabled there.
Watch for this againA new watcher from Scout's proposed rule, opened in the watcher editor. It starts in Shadow, so you see matches before it acts. When it fires names who gets told, or warns that nobody will be notified. See Watchers.
NotifyOff unless you tick it. After concluding, it opens an email in your mail app with the verdict, your reason and a link to the hunt, for you to address and send. Huntbase doesn't send it.

Conclude​

At the bottom of the sheet, click Conclude hunt. To record the verdict but keep working, tick Keep hunt open; the button then reads Record verdict. On a hunt that's already concluded, it reads Update conclusion.

Beside the button, a line says what will be recorded and on which evidence, for example Records “Clear” on 2 checks with results, 1 check that returned no rows, 1 other step as evidence and concludes the hunt. Cited cells are counted three ways: checks with results (they returned rows), checks that returned no rows, and other steps, such as an analytic or a task, which have no row count. If the line doesn't match what you meant to cite, go back to Evidence before you conclude.

The sheet then shows Concluded · and the verdict. Once the hunt's finding is saved it says Saved as a finding: the report you can print or share. and its main button is Open finding. Below that, Next steps lists the follow-ups you chose (Create the watcher (starts in Shadow), Email the team, the containment links). Concluding also clears the hunt's progress messages, such as Query regenerated — ready to run.

If the finding can't be saved, the verdict still stands: the sheet says the finding wasn't saved and offers Retry.

The finding​

Concluding a hunt saves a finding: the record of what the hunt found, and the report you print or share. Concluding again updates the same finding with the new verdict; customer wording you wrote earlier is kept. A finding holds:

  • the verdict, severity, your reason and who concluded the hunt, and when;
  • the evidence cells, each linked to its cell in the hunt, with its row and tagged-row counts;
  • the entities the hunt involved, by role;
  • the follow-ups you chose, each with a status (To do, Done, Failed, Skipped) you can update on the finding;
  • your Notes for the team;
  • the customer copy, built from the hunt.

Open it with Open finding, from the hunt's Linked strip (Finding · and the verdict), or from the Findings list on Hunt reports. See Findings for the finding page, printing and share links.

The verdict cell​

The Verdict cell sits last in the notebook. It is a summary, not a form:

  • the headline is your recorded reason, or, before you conclude, how the hunt's cells stand (Nothing has run yet. The verdict fills in as cells finish.);
  • the verdict and any acknowledged issues, for example Confirmed · 2 open issues acknowledged;
  • before you conclude, how many open issues Conclude will list, and Scout's draft if there is one (Scout drafted Benign. Review it in Conclude: nothing is recorded until you do.);
  • Conclude or Edit conclusion, and Open report or Generate report. Open report stays available on a concluded hunt, for anyone who can see it;
  • the Verdict log: every verdict recorded, marked Draft (Scout's), Recorded or Superseded, with its reason, acknowledged issues and follow-ups.

What concluding does​

Concluding closes the hunt and saves its finding. A concluded hunt is read-only until someone who can edit it reopens it, and its status chip reads Concluded · with the verdict. Closing settles work still in flight: steps waiting for approval are marked Not run (hunt closed) and don't count as failed. Running queries and collections stop, and the hunt drops out of Needs you and its counts. Notes and endpoint cells are left as they are, and Scout stops working on the hunt: it doesn't add cells to a concluded hunt. If you ask it for more, it tells you what it would check next instead. To let Scout add to the hunt again, reopen it.

Reopen, archive or delete​

ActionWhereWhat happens
Reopen hunt⋯ on a concluded hunt, or the read-only stripThe hunt is active again. The verdicts it was concluded on are marked Superseded, and the verdict cell says hunt reopened — conclude again. You can run the stopped and not-run steps again.
Archive hunt⋯ on an active huntShelves the hunt. It stays readable.
Restore to active⋯ on an archived huntBrings it back.
Delete draft…⋯ on a draft, owner onlyDeletes the draft after a confirmation.

These are open to the owner and anyone with Can edit, except Delete draft…, which only the owner can use.

Save, export and reuse a hunt​

Item in ⋯What it does
Save to Library › Save as playbook…Saves the hunt as a hunt playbook you can launch again.
Save to Library › Relaunch when a watcher fires…Arms a trigger that relaunches this hunt's playbook. See below.
Export › Customer report…The customer report, once a verdict is recorded. See Customer or shareable report.
Export › Hunt as YAML…Downloads the hunt as YAML. Tick Flow only for a clean, reusable structure.
Export › Results spreadsheet…The incident tracker workbook (.xlsx). See Tracker spreadsheet.
Import cells from YAML…Adds cells from a YAML export to this hunt.

The exports stay available on a concluded hunt.

Relaunch automatically when a watcher fires​

Open ⋯ › Save to Library › Relaunch when a watcher fires… to Arm a trigger: pick a watcher pattern, a session mode and a severity floor, and Huntbase launches this hunt's playbook whenever that watcher fires. If the hunt is not a playbook yet, arming saves it as one first. Armed triggers are listed with their state (Armed, Needs data, Backed off) with Run now and Remove. See Watchers.

Write the report​

Generate report has Scout write the hunt up against the template you pick (one of your organisation's, one from a content pack, or one Huntbase ships) and Open report opens it as its own tab. The report draws on the hunt's results, entities, tags, notes and investigation timeline. Caveats such as a draft report on an open hunt, or claims the evidence doesn't support, are shown above the body. Download saves it as a Word document (.docx) or Markdown (.md), and Regenerate rewrites it after the hunt has moved on. See Hunt reports.

For a customer or anyone outside the security team, use Customer report (or Shareable report). See Customer or shareable report.

Next steps​