Skip to main content

Findings

A finding is one vulnerability in one installed package on one asset. A container image that runs in several pods is one asset, so its findings are listed once, with every place it runs. The Findings page lists them all, and is where you decide what to fix first.

The findings table with its filters

The findings table with its filters

Reading the table​

ColumnShows
SeverityThe severity level, from critical to unknown. Select the heading to sort.
CVSSThe base score, when the data has one.
VulnerabilityThe identifier, such as a CVE- or GHSA- number. It links to the entry of the vulnerability at its source, in a new tab.
PackageThe package name, its installed version and its type.
FixThe fix state: the version to upgrade to, or why there is none.
LocationWhere the package is: the host, or the cluster, namespace, pod, container and image. Select it to see the dependencies of that host or container.
First seenThe first scan that reported this finding.

The Fix column tells you the next step. Fixed shows the version to upgrade to. See Key concepts for each state.

Under the identifier, two signs help you rank the work:

  • Exploited (KEV): CISA lists the CVE in its Known Exploited Vulnerabilities catalog. Fix these first.
  • EPSS: the chance of exploitation in the next 30 days, from FIRST. Sort by EPSS to see the likeliest exploits first. The scores are refreshed once a day.

Triage​

Users with the findings.triage permission see a Triage button on each row. Use it to record what you decided:

StatusMeaning
OpenNobody has decided yet. Use it to set only an owner or a ticket.
Under investigationSomeone is looking at it.
AffectedThe package is affected, and the note says what to do.
FixedThe fix is in place, but the scan has not seen it yet.
Not affectedThe vulnerable code is not in use. Pick the justification.

A Not affected finding leaves the list, the exports and the alerts. Its decision stays on the server, so the next scan does not bring it back. To show the decisions again, remove the decision on the API; see the API reference.

Each decision can also name an owner and a ticket link. The server does not open tickets for you: create the ticket in your tracker, then paste its link.

Filtering​

The filter bar works like the one on the dashboard. Use it to answer questions such as:

  • Which critical findings in production can be fixed today? Set Severity to critical, Cluster to the production cluster, and select Fixable only.
  • What is new this week? Set First seen since to seven days ago.
  • Where is log4j? Enter log4j in Package, or search for a CVE in CVE.
  • What does this host carry? Set Kind to hosts and Hostname to the host name.

Search looks at identifiers, package names and pod names at once. Use it for a quick lookup; use the dedicated fields when you need exact matches.

Sorting and paging​

Select a column heading to sort by it, and select it again to reverse the order. The table is split into pages; the number of matching findings is shown above it.

Empty results​

  • No findings yet: no agent has sent an inventory, or no scan has run since. Check the Assets page for Scan pending.
  • No findings match these filters: remove a filter, or select Clear filters.

Export​

CSV, JSON, SARIF and PDF export the findings that match the filters. The files keep the same filters, so you can hand over exactly the list you see.

SARIF feeds CI security tabs, such as the code scanning of GitHub. Each finding is a result with its severity as the level. OpenVEX exports the triage decisions instead, so other tools can read which findings are not affected, fixed or under investigation. For an export that you can repeat later, create a report from the same filters; see Reports.

Next steps​

Select a location to open the Dependencies of that asset, and from there the package with its full list of assets and findings.