AWS guardduty: Document Runtime Monitoring Custom Detection Rule finding details
Summary
Restructures the Custom Detection Rules findings section: the existing `Service.Detection.Sequence` description is now scoped to AWS CloudTrail rules, and a new 'Findings from Runtime Monitoring rules' subsection documents the Type format (`Tactic:Runtime/Technique`), the `CustomDetectionRuleName` field, `Service.RuntimeDetails.Context` fields (FilePath, FileOperation, CommandLineExample, ServiceName, ToolName, ToolCategory), and per-resource finding aggregation.
Security assessment
The change is documentation of the schema/fields of GuardDuty Runtime Monitoring Custom Detection Rule findings (finding type format, rule name, runtime context fields). It improves understanding of detection output but does not describe a vulnerability, patch, credential handling, or access-control change, so severity is low.
Evidence
+Findings from Runtime Monitoring rules use the same structure as other Runtime Monitoring findings, including the runtime details of the process involved. For more information, see Runtime Monitoring finding details. The following details are specific to Custom Detection Rules:
Diff
diff --git a/guardduty/latest/ug/guardduty_findings-summary.md b/guardduty/latest/ug/guardduty_findings-summary.md index 547a382..42cf6db 100644 --- a//guardduty/latest/ug/guardduty_findings-summary.md +++ b//guardduty/latest/ug/guardduty_findings-summary.md @@ -516 +516,5 @@ Network indicator value | Description -Custom Detection Rules findings use the same `Service.Detection.Sequence` structure as Attack sequence finding details. The Actors, Endpoints, Resources, and Signals fields share the same schema. This section describes the differences specific to Custom Detection Rules findings. +The details in a Custom Detection Rules finding depend on the data source of the rule that generated it. + +### Findings from AWS CloudTrail rules + +Findings from AWS CloudTrail rules use the same `Service.Detection.Sequence` structure as Attack sequence finding details. The Actors, Endpoints, Resources, and Signals fields share the same schema. The following details are specific to Custom Detection Rules. @@ -542,0 +547,33 @@ Details about the network endpoints from which the activity originated. Includes +### Findings from Runtime Monitoring rules + +Findings from Runtime Monitoring rules use the same structure as other Runtime Monitoring findings, including the runtime details of the process involved. For more information, see Runtime Monitoring finding details. The following details are specific to Custom Detection Rules: + +**Type** + + +The finding type follows the format `Tactic:Runtime/Technique`. For example, `Persistence:Runtime/SuspiciousCommand`. + +**Rule** + + +The `CustomDetectionRuleName` field contains the name of the Custom Detection Rule that generated the finding. + +**Context** + + +The `Service.RuntimeDetails.Context` fields describe the matched activity. The fields that a finding includes depend on the rule: + + * Rules that generate `SensitiveFileRead` and `SensitiveFileModified` findings include `FilePath` and `FileOperation`. + + * Most rules that generate `SuspiciousCommand` findings include `CommandLineExample`, and some also include `ServiceName`. Rules that detect activity on a specific file, such as changing file permissions, include `FilePath` instead, and `FileOperation` when it applies. + + * Rules that generate `SuspiciousTool` findings include `ToolName` and `ToolCategory`. + + + + +**Resource** + + +Each finding covers a single resource, such as an instance, a task, or a container. For more information about how GuardDuty groups signals, see [Finding aggregation](./custom-detection-rules-how-it-works.html#custom-detection-rules-how-it-works-finding-aggregation). +