Implementing Diamond Model Analysis

Implementing Diamond Model Analysis

Overview

The Diamond Model of Intrusion Analysis provides a structured framework for analyzing cyber intrusions by examining four core features: Adversary, Capability, Infrastructure, and Victim. This skill covers implementing the Diamond Model programmatically to classify and correlate intrusion events, build activity threads linking related events, create activity-attack graphs, and generate pivot-ready intelligence from intrusion data.

When to Use

• When deploying or configuring implementing diamond model analysis capabilities in your environment
• When establishing security controls aligned to compliance requirements
• When building or improving security architecture for this domain
• When conducting security assessments that require this implementation

Common Misconfigurations & Verification

• Empty core features break pivots: events missing one of the four vertices (Adversary/Capability/Infrastructure/Victim) silently drop out of find_pivots(), which only pivots on non-empty fields. Validate every DiamondEvent populates all four before adding it.
• Unnormalized infrastructure values: 1.2.3.4, 1.2.3.4:443, and hxxp://1.2.3.4 are treated as distinct infrastructure nodes, so shared-infra pivots are missed. Canonicalize values before graphing.
• Timestamp format drift: build_activity_thread() sorts lexically on the timestamp string - mixed formats (epoch vs ISO-8601) corrupt thread ordering. Enforce UTC ISO-8601.
• Capability not mapped to ATT&CK: leaving mitre_techniques empty prevents cross-event capability correlation and ATT&CK-based clustering of activity groups.
• Over-merging activity groups: clustering threads on shared infrastructure alone (CDN, shared host) creates false activity groups - require a capability or adversary pivot as well.
• Verification: confirm pivot output groups only events with >1 shared value, render the activity-attack graph and check edges are chronological, and verify each event resolves to valid ATT&CK technique IDs.

Prerequisites

• Python 3.9+ with networkx, stix2, graphviz libraries
• Understanding of the Diamond Model core and meta-features
• Access to threat intelligence data (MISP/OpenCTI events)
• Familiarity with MITRE ATT&CK for capability mapping

Key Concepts

Diamond Model Core Features

• Adversary: The threat actor or operator conducting the intrusion
• Capability: The tools, techniques, and malware used (maps to ATT&CK)
• Infrastructure: C2 servers, domains, email addresses, hosting providers
• Victim: Target organization, system, person, or data asset

Meta-Features

• Timestamp: When the event occurred
• Phase: Kill chain stage (recon, delivery, exploitation, etc.)
• Result: Success, failure, or unknown
• Direction: Adversary-to-infrastructure, infrastructure-to-victim, etc.
• Methodology: Social engineering, technical exploit, insider threat
• Resources: Financial, human, technical resources required

Activity Threads and Groups

• Activity Thread: Sequence of Diamond events from a single adversary operation
• Activity Group: Cluster of threads attributed to the same adversary

Workflow

Step 1: Define Diamond Event Data Structure

from dataclasses import dataclass, field
from datetime import datetime
from typing import Optional
import json
import uuid

@dataclass
class DiamondEvent:
    adversary: str = ""
    capability: str = ""
    infrastructure: str = ""
    victim: str = ""
    timestamp: str = ""
    phase: str = ""
    result: str = ""
    direction: str = ""
    methodology: str = ""
    confidence: int = 0
    notes: str = ""
    event_id: str = field(default_factory=lambda: str(uuid.uuid4())[:8])
    mitre_techniques: list = field(default_factory=list)
    iocs: list = field(default_factory=list)

    def to_dict(self):
        return {
            "event_id": self.event_id,
            "adversary": self.adversary,
            "capability": self.capability,
            "infrastructure": self.infrastructure,
            "victim": self.victim,
            "timestamp": self.timestamp,
            "phase": self.phase,
            "result": self.result,
            "direction": self.direction,
            "methodology": self.methodology,
            "confidence": self.confidence,
            "mitre_techniques": self.mitre_techniques,
            "iocs": self.iocs,
            "notes": self.notes,
        }

Step 2: Build Activity Thread from Events

import networkx as nx

class DiamondAnalysis:
    def __init__(self):
        self.events = []
        self.graph = nx.DiGraph()

    def add_event(self, event: DiamondEvent):
        self.events.append(event)
        self.graph.add_node(event.event_id, **event.to_dict())

    def build_activity_thread(self):
        """Link events chronologically into activity threads."""
        sorted_events = sorted(self.events, key=lambda e: e.timestamp)
        for i in range(len(sorted_events) - 1):
            self.graph.add_edge(
                sorted_events[i].event_id,
                sorted_events[i + 1].event_id,
                relationship="followed_by",
            )

    def find_pivots(self):
        """Find pivot points where events share infrastructure or capabilities."""
        pivots = {"infrastructure": {}, "capability": {}, "adversary": {}}

        for event in self.events:
            if event.infrastructure:
                pivots["infrastructure"].setdefault(event.infrastructure, []).append(event.event_id)
            if event.capability:
                pivots["capability"].setdefault(event.capability, []).append(event.event_id)
            if event.adversary:
                pivots["adversary"].setdefault(event.adversary, []).append(event.event_id)

        return {
            k: {pk: pv for pk, pv in v.items() if len(pv) > 1}
            for k, v in pivots.items()
        }

    def generate_report(self):
        return {
            "total_events": len(self.events),
            "unique_adversaries": len(set(e.adversary for e in self.events if e.adversary)),
            "unique_victims": len(set(e.victim for e in self.events if e.victim)),
            "unique_infrastructure": len(set(e.infrastructure for e in self.events if e.infrastructure)),
            "pivots": self.find_pivots(),
            "events": [e.to_dict() for e in self.events],
        }

Validation Criteria

• Diamond events capture all four core features with meta-features
• Activity threads link related events chronologically
• Pivot analysis identifies shared infrastructure and capabilities across events
• Graph visualization renders the activity-attack graph correctly
• Events map to MITRE ATT&CK techniques for capability classification

References

• Diamond Model Paper
• MITRE ATT&CK
• STIX 2.1 Campaign Object