DocumentationProvenance & Attribution

About the Unified MITRE Navigator

Understanding the data sources, ingestion methodology, edge confidence classifications, and copyright attributions powering this cross-framework cyber knowledge graph.

Integrated MITRE Knowledge Bases

The Navigator harmonizes five distinct MITRE cybersecurity repositories into a single relational graph:

Offensive BehaviorsOfficial Site ↗

MITRE ATT&CK® (Enterprise & ICS)

Globally-accessible knowledge base of adversary tactics, techniques, and procedures (TTPs) based on real-world observations.

Source Format:STIX 2.1 JSON bundles (enterprise-attack.json, ics-attack.json)
License:MITRE Terms of Use / CC BY 4.0
AI & ML Threat MatrixOfficial Site ↗

MITRE ATLAS™

Adversarial Threat Landscape for Artificial-Intelligence Systems — cataloguing vulnerabilities and attack patterns targeting ML models.

Source Format:STIX JSON bundle (stix-atlas.json)
License:MITRE Terms of Use / CC BY 4.0
Attack MechanismsOfficial Site ↗

MITRE CAPEC™

Common Attack Pattern Enumeration and Classification — comprehensive dictionary of known patterns of attack employed by adversaries to exploit weaknesses.

Source Format:CAPEC-3 XML schema catalog (capec_attack_pattern.xml)
License:MITRE Terms of Use / Copyright © The MITRE Corporation
Defensive CountermeasuresOfficial Site ↗

MITRE D3FEND™

Knowledge graph of cybersecurity countermeasure techniques and digital artifact relationships designed to tailor defensive capabilities.

Source Format:SPARQL Knowledge Graph Results (defend-full-mappings.json)
License:MITRE Terms of Use / CC BY 4.0
Deception & DenialOfficial Site ↗

MITRE Engage™

Framework for planning and discussing adversary engagement, cyber deception, and denial activities aligned with ATT&CK.

Source Format:Raw Activity & Crosswalk Mappings (activities.json, attack_mapping.json)
License:MITRE Terms of Use / CC BY 4.0
Relationship Methodology

Official Crosswalks vs. Inferred Semantic Links

To maintain strict analytical rigor, all connections in the knowledge graph are categorized by confidence:

Official MITRE Crosswalks

Direct cross-framework mappings published and maintained by MITRE:

  • ATLAS → ATT&CK: Referenced via STIX external_references.
  • CAPEC → ATT&CK: Explicitly declared via Taxonomy_Mappings in CAPEC-3.
  • D3FEND → ATT&CK: Extracted from D3FEND OWL ontology SPARQL query bindings (off_tech_id).
  • Engage → ATT&CK: Sourced from official MITRE Engage crosswalk tables (attack_mapping.json).

Inferred Semantic Links

Heuristic connections surfaced via automated natural language processing (TF-IDF keyword extraction & Jaccard similarity indexing):

  • Identifies latent conceptual connections for techniques lacking official crosswalk records.
  • Rendered with dashed lines in the canvas and marked with distinct inferred badges.
  • Can be toggled off at any time via the filter bar to isolate purely official MITRE relationships.

Data Provenance & Prebuild Pipeline

The ingestion pipeline is executed at prebuild time via scripts/ingest.ts, producing a compact, static graph.json bundle with zero runtime backend dependencies.

FrameworkRaw Source FileObjects IngestedMapping Relationship
ATT&CK Enterpriseenterprise-attack.jsonTop-level techniques (non-subtechniques)Primary Anchor Taxonomy
ATT&CK ICSics-attack.jsonIndustrial Control System techniquesICS Anchor Columns
ATLASstix-atlas.jsonAI / ML attack patternsanalogue-of → ATT&CK
CAPECcapec_attack_pattern.xmlStandard & Meta attack patternsmaps-to → ATT&CK
D3FENDdefend-full-mappings.jsonDefensive countermeasuresmitigates → ATT&CK
Engageactivities.jsonEngagement activitiescounters → ATT&CK