Skip to main content

Entities

Entities are the things your data keeps talking about: hosts, users, IP addresses, domains, URLs, email addresses, files, hashes and processes discovered in query results, plus vulnerabilities, ATT&CK techniques and other public threat-intelligence objects Huntbase already knows about. Wherever one appears — a result table, a Scout answer, a report, an insight — it behaves the same way: hover it for a quick read, inspect it in a side panel, or open its full profile.

Entities open in Activity with Type chips and counts, one row selected

Entity chips​

Values Huntbase recognises show as entity chips: an icon for the kind, the value, and a coloured dot when threat intel has a verdict on it. Chips appear in hunt results tables, in Scout's answers, in notebook notes and in reports. A user is shown as a chip, never as an email link.

Huntbase decides what a value is from both its shape and its column. A column named src_ip or ip_address holds IP addresses; a description, message or user_agent column is never treated as one. Hosts, users, processes and accounts are only recognised in a column that says what they are, so IDs, timestamps and bare numbers don't turn into chips.

ToDo this
Get a quick readHover or focus the chip for a moment, or press Space.
Inspect itClick the chip or press Enter. Where there's no side panel to open, the click pins the card open instead.
Open its profile⌘-click (Ctrl-click on Windows/Linux).

A violet ring around a chip means the value is already part of the hunt you're in.

The hover card​

The hover card answers "is this bad, and have we seen it?" without leaving the page:

  • Open profile in the header.
  • A verdict line from threat intel, for example "Malicious · Katana · ThreatFox 75%". Intel that has expired or been revoked reads as history, for example "Listed by MISP (demo) · expired 22 Oct — treat as historical". When the lookup fails, the card says Intel lookup unavailable. No intel never reads as "clean".
  • Up to four facts (source, confidence, first and last seen).
  • How many times it was seen in your estate, or Not seen in your estate.
  • Actions: Inspect, Ask Scout, Add to hunt and Copy (Copy defanged for IP addresses, domains, URLs and email addresses, such as 193.233.20[.]122). An action the current page can't carry out is left out.

Right-click a table cell​

Right-click a host, user, IP address, process, hash, domain, URL or email cell in a results table for the entity's actions at the top of the menu, above Add to timeline, Tag and Note: Ask Scout, Search for this, Open in graph, Open endpoint (hosts, when the host is in one of your fleets) and Copy.

Where to find entities​

Entities live in Activity but don't have a tab of their own — there are usually far too many. Open them in one of these ways:

  • Click or ⌘-click any entity chip.
  • Press ⌘K (Ctrl+K on Windows/Linux) and choose Browse entities. When what you type names an entity, the palette shows an Open … row for it.
  • Follow an old /entities bookmark — it forwards to the entities list.

In the list, the header search box (Search activity…) narrows the rows. Search is a case-insensitive substring match on the entity's name or value; 192.168. finds every address in that range, and srv-web finds srv-web-01. There are no wildcards.

ControlWhat it does
Type chipsOne chip per entity type present in your data, with a count — for example endpoint, user, IP address, domain, URL, file, process, CVE, ATT&CK technique or indicator. Platform administrators can add types.
Time rangeThe same picker as the rest of the feed — presets or a custom range.
SortNewest first, Oldest first, Most severe first, Least severe first, Name A–Z, Name Z–A.
Per page10, 20, 30, 50 or 100 rows.

Inspect an entity​

Click an entity row or chip to open it in the side panel. The panel always answers the same questions in the same order: is it bad, what can I do, and what do we know.

Entity side panel with the verdict strip, action row and key facts

Header. The value, its type and context (for example Endpoint · macOS), a source badge (Public, Private, Manual, Agent or System), an Inactive badge where the source has retired it, and an Open profile link (⌘/Ctrl+Enter).

Verdict strip. For IP addresses, domains, URLs, hashes and email addresses, the verdict comes from threat intel, for example "Malicious · Katana botnet C2" with the source, confidence and how often it was seen in your estate, and an Open indicator link to the indicator. A value nothing knows about reads "Nothing known about this IP". If the lookup fails, the strip says Intel lookup unavailable with a retry. For hosts, users and other kinds, the strip shows the open work on it ("2 findings · 1 hunt", or No open findings or hunts).

Action row. One row, always in this order, showing only what applies:

ActionShortcutWhat it does
Ask ScoutSAsks Scout about the entity, in the current hunt or a chat.
Add to huntHNew hunt from this… opens Hunt with Scout with the entity as the seed. See Start a hunt from an entity.
Look backLSearches your history for the value. IP addresses, domains, URLs, hashes and email addresses only. See Indicators.
Watch for thisWAdds the value to a watch list so future sightings are flagged. Same kinds as Look back.
RespondResponse actions… on the endpoint's page. Endpoints in your fleets only.
⋯Open endpoint, Open in graph, Add to timeline…, Export as JSON / Export as CSV, Copy value and Copy defanged.

Sections, in one scroll:

SectionContents
Key factsThe handful of attributes that matter for this type. Left out when there are none.
In this huntWhen you inspect from a hunt: where the entity appears in it.
Summary stripEndpoint agent with its status (or No endpoint agent), Last seen, Detections on the entity, the hunt about it, and Also known as — other records of the same thing.
Seen onThe endpoints and users the value was seen on, with counts, the hunts it appeared in, and the first and last time seen in your estate.
Related intelThe malware, actors and labels of the matching indicator.
Scout summaryShown when Scout has written one. To get a fresh read, use Ask Scout.
ActivityCollapsed. Opens the entity's timeline in place, with Open full page — see Entity timeline. Hosts, users, IP addresses and file hashes.
Tags / NotesTags (including ones Scout suggested) and your team's notes. See Tags and notes.
DetailsCollapsed. Raw IDs, the full attribute set, and where the entity came from (Found by run or From query).

Pivot without losing your place​

Click a chip inside the panel — an endpoint under Seen on, a value under Key facts — and the panel moves to that entity instead of opening a new tab. A breadcrumb at the top shows the trail; click any step to go back, or press Backspace or [ to go back one. The trail is in the page URL, so a shared link opens the same stack. Closing the panel clears it.

When Huntbase has no record​

Some values are seen in your data without a saved record behind them. Instead of a dead end, the panel says "Huntbase has no profile for this user yet. It was seen in your data, but no saved record backs it." and offers Search for this and Open in graph.

The entity profile​

Open profile — on a chip, a hover card or the side panel — opens the entity's own page. Profiles exist for endpoints, users, IP addresses, domains, URLs, email addresses, hashes and files. The page title and breadcrumb name the kind ("IP address profile"), never the value.

The profile lays out the same answers at full width:

  • The verdict strip from threat intel, and a security state line from open work: for example "At risk · 1 Critical hunt open", or "No open hunts or insights mention this endpoint" when nothing is open.
  • For endpoints, a separate agent line: "Endpoint agent · online · Agent healthy (the agent's own health, not the host's security)", with a link to the Endpoint page.
  • The action row, with Respond for endpoints in your fleets.
  • Open work — hunts and insights from the last 90 days, open first, with severity and why each is listed: In its results, Observed as an IOC, Linked to it or Names it (its title names the entity). Hunts linked only by hand may be missed.
  • Activity — Open its timeline, the endpoint page for activity, logs and live response, and Search telemetry for this.
  • Key facts, Seen on, Related intel and Also known as.

Open in graph​

⋯ › Open in graph draws the entity's relationships in an Explorer graph tab — the entity in the centre, its neighbours around it, edges labelled with the relationship (for example Detected on).

  • Click a node to inspect it.
  • Right-click a node for Expand (All relationships or a suggested relationship), Inspect, Ask Scout, Add to hunt › New hunt from this, Search for this, Open endpoint (device nodes) and Copy. A restricted node offers Inspect only.
  • Filters hides or shows nodes by Node types and edges by Relationships, each with a count.
  • Ask Scout in the graph toolbar starts a chat about what is on the canvas. The chat shows a collapsed Canvas shared with Scout chip with the entity and relationship counts.

More than ten neighbours of the same type hanging off one node are drawn as a single collection node badged with how many (82 insights). Click it to see every member, or use Group in the toolbar to turn folding off. Only neighbours whose only connection is that node are folded, and the results table still lists every entity. The graph has no time filter; it shows relationships as they stand now.

Relationships graph seeded on an entity, right-click menu open on a neighbour

Start a hunt from an entity​

Rolling out

Hunts are rolling out and may not be enabled for your organization yet.

Add to hunt › New hunt from this… appears on entities where a hunt makes sense — adversary behaviour (techniques, groups, campaigns, software), vulnerabilities and CVEs, indicators and observables (IPs, domains, URLs, files and hashes, malware) and your own assets (endpoints, users, accounts). Reference material such as CWE, CPE and packages doesn't offer it.

It opens Hunt with Scout with the entity as the seed. Type What are you looking for? (optional, for example "did this run on our edge devices in the last 30 days?"), adjust the Starting hypothesis Scout drafts, and click Start with Scout. See Start a hunt.

How entities are created​

You rarely create entities by hand. They arrive from:

  • Query results — Huntbase extracts entities from the rows a query returns and records the run, query and row they came from (see Details). These carry the Private source badge.
  • Endpoint Control and connections — endpoints and other objects synchronised from your integrations. See Endpoint Control and Connections.
  • Public threat intelligence — CVEs, ATT&CK techniques and similar objects that Huntbase maintains for everyone. These carry the Public badge.
  • Scout — Scout creates entities during investigations (Agent).

Next steps​

  • Entity timeline — everything about one host, user, IP or hash, in order
  • Indicators — the threat intel behind an entity's verdict
  • Insights — the detections that reference these entities
  • Start a hunt — what happens after New hunt from this…