Skip to main content

Worked Examples: CTI Foundations

Purpose

Show what acceptable requirement, source, evidence, and confidence artifacts look like before analysis begins.

Bad PIR vs Corrected PIR

TypeExampleProblem / Strength
Bad PIR"Tell us everything about MuddyWater."Topic, not a decision. No scope, consumer, time window, or output.
Corrected PIR"Which reported MuddyWater behaviors since 2024 are observable with our endpoint, email, and identity telemetry, and should they enter a 30-day hunt backlog?"Decision-linked, bounded, telemetry-aware, and actionable.

PIR to SIR to EEI Decomposition

Decision owner: SOC Lead
Decision: Approve or reject a scoped 30-day hunt for RMM abuse.

PIR-001:
Which publicly reported RMM-abuse behaviors are relevant to our environment?

SIR-001:
Which actors or campaigns are reported to use RMM tools after initial access?

EEIs:
- Source title, publisher, publication date, accessed date
- Tool names
- Initial access vector
- Process behavior
- Network behavior
- Reported target sectors
- Evidence label and confidence

SIR-002:
Can the SOC observe these behaviors?

EEIs:
- EDR process creation fields
- Software inventory fields
- Network destination fields
- Identity session fields
- Known approved RMM baseline

Source Register Row

FieldFilled Example
source_idSRC-2026-001
publisherMITRE ATT&CK
titleEnterprise Matrix / Technique documentation
urlhttps://attack.mitre.org/
publication_dateOngoing
accessed_date2026-05-16
source_typeFramework documentation
source_reliabilityA
information_credibility2
limitationsTechnique pages are not campaign proof or attribution evidence.
downstream_useATT&CK mapping quality gates and detection backlog rules.

Evidence Register Row

FieldFilled Example
evidence_idEV-FOUND-001
claimATT&CK is a behavior taxonomy and should not be used as attribution proof.
evidence_labelReported
source_idSRC-2026-001
confidenceHigh confidence
confidence_reasonPrimary framework source and consistent with defensive practice.
contradiction_or_gapNone for doctrine use; campaign use still requires separate evidence.
downstream_useATT&CK mapping pages, detection backlog, attribution rules.

Analyst Checklist

  • Is the requirement a decision question?
  • Does every source have publication/access metadata?
  • Does every material claim have an evidence label?
  • Is confidence justified by evidence quality, not analyst preference?
  • Are gaps actionable or explicitly accepted?