HERA | Intelligence & Analysis Platform

System Overview

A connected environment for evidence, entities, and analysis.

HERA is built as three connected layers rather than a single tool: an evidence layer for intake and normalization of source material, a graph layer that represents entities and their relationships as explicit, queryable structure, and a workflow layer that carries research from an open question to a structured, reviewable finding.

Every item that enters HERA is scoped to a case. Documents, records, and structured data are preserved alongside their original source, so any entity profile, relationship, or conclusion can be traced back to the material it was built from. HERA is currently in active development: the evidence data model and graph exploration interface are running in an internal prototype, and the workflow tooling around them is being built out.

Domain: Intelligence / Analysis
Status: Active development
Core loop: Intake → Link → Analyze → Report
Primary users: Analysts, researchers & investigative teams

Evidence & Entity Workflows

From raw material to structured, linked records.

Every piece of material entering a case runs through the same structured pipeline, regardless of whether it started as a document, a dataset, a media file, or a manual entry.

01

Intake

Documents, records, media, and structured exports are added to a case workspace, each preserved with its original source and intake metadata.

02

Normalize

Formats, dates, identifiers, and naming conventions are reconciled into a consistent internal representation so material from different sources can be compared directly.

03

Extract Entities

People, organizations, locations, events, and objects referenced in the material are identified and represented as discrete, attributable entities.

04

Link & Deduplicate

Candidate matches across records are surfaced for review, and confirmed links merge fragmented references into a single entity record.

05

Review & Annotate

Analysts confirm, annotate, and flag findings, with every judgment attached to the evidence it was drawn from.

Graph Exploration Model

Relationships as explicit, navigable structure.

Entities and the relationships between them are modeled explicitly in HERA, not inferred at read time from unstructured text. An analyst can start from any single record and move outward through its connections.

Starting from a person, organization, or location, an analyst can expand outward through the relationship types that produced each link — shared address, shared communication, shared transaction, or co-occurrence in the same document — rather than reconstructing those connections by cross-referencing files manually.

Graph views can be filtered by relationship type, time range, source, and evidentiary strength, and specific views can be saved and reused across similar cases, so relationship structure becomes something an analyst inspects directly rather than something held in memory.

Entity Types

Person
Organization
Location
Event
Communication
Asset
Document
Account

Relationship Types

Associated With
Communicated With
Located At
Owns / Controls
Participated In
Referenced By
Transacted With
Co-Occurs With

Recon & Analysis Workflows

Structured research passes, not open-ended searching.

A “recon” pass in HERA is a structured research pass over case material and available reference data: establishing what is known, tracking what remains open, and generating leads for the next pass, with the process itself kept auditable.

01

Source Coverage Tracking

Tracks which sources, registries, and records have been checked for a given entity or question, and which remain open, so research effort stays visible and repeatable.

02

Cross-Referencing & Verification

Compares claims and records across independent sources to confirm, qualify, or flag inconsistencies before they become findings.

03

Lead & Hypothesis Tracking

Captures open questions and working hypotheses as structured items linked to the evidence that raised them, rather than losing them in notes.

04

Timeline Reconstruction

Assembles dated events and records from multiple sources into a single ordered sequence, to test overlap, sequence, and consistency.

Situational Insight Outputs

Case state turned into artifacts analysts can act on.

HERA's outputs are built to be used, shared, or handed off — direct products of the current graph and workflow state, not a separate reporting step bolted on afterward.

Case Brief

A structured summary of a case's entities, relationships, and open questions, generated from the current state of the graph and workflow.

Entity Dossier

A consolidated profile for a single entity: attributes, linked records, relationship history, and source provenance in one view.

Relationship Map Export

A shareable export of a filtered graph view, built for briefings or handoff to another analyst.

Change Notification

An alert when new material affects an entity or relationship already under analysis, so re-review happens deliberately rather than by accident.

Development Status

Active development, built layer by layer.

HERA is under active development. The evidence data model, ingestion pipeline, and graph exploration interface are functional in an internal prototype; workflow tooling, review interfaces, and structured export are being built out around that foundation.

Case & Evidence Data Model

Implemented

The case-scoped data model for evidence, entities, and relationships is in place and used by every other layer of the system.

Ingestion & Normalization Pipeline

In Progress

Intake and normalization for documents and structured data exports is operational; additional source formats are being added.

Graph Exploration Interface

Prototype

Entity and relationship browsing, filtering, and saved views are running in an internal prototype interface.

Workflow & Annotation Tooling

Early Stage

Review, annotation, and lead-tracking tools are being designed around the workflows analysts actually use during a case.

Structured Export & Reporting

Planned

Case briefs, entity dossiers, and relationship exports are the next milestone once workflow tooling stabilizes.

Note Interface captures and system diagrams will be published here as the prototype matures.

Collaboration

Technical collaboration on structured analysis and investigative tooling.

MDNT is open to focused collaboration on HERA: evaluation environments, data model and workflow feedback, integration scenarios, and partners with real investigative, research, or analytical workflows to pressure-test the platform against. If your work touches structured analysis, evidence review, or investigative research, get in touch.

Evaluation environments for analysts & research teams
Data model & workflow feedback
Integration with existing case management or research tools
Early access & pilot collaboration

labs@mdnt.pt

PGP encrypted contact (optional)

Fingerprint: 20F4 1435 3E7E A1A6 7EC0 EA11 8DFF F119 FF5B 8464
Algorithm: ECC (Curve25519)
Function: Encryption, decryption, signing, verification
Created: 2026-07-10

Public Key

Structured analysis for evidence, entities, and investigative work.