SYSTEM ARCHITECTURE // V2.1

CONFLICT INTELLIGENCE ENGINE

A multi-pipeline, multi-agent Conflict Intelligence Engine. Tacitus fuses OSINT scraping, deep-document RAG, and a full conflict ontology to build a living model of political, organizational, and interpersonal disputes.

01. SYSTEM TOPOLOGY

>> The Intelligence Pipeline

The Concordia Engine operates across a structured analytical chain: Ingestion → Agentic Compute → Strategic Output. Each layer is fault-tolerant, independently scalable, and designed to model conflicts ranging from family disputes to geopolitical crises.

INPUT LAYER

DATA INFRASTRUCTURE

  • > Tavily / Serper / GDELT
  • > Firecrawl Local Media Scraping
  • > RAG (UN, ICG, SG Reports)
  • > Supabase Vector Search
  • > Vertex AI Batch Inference
COMPUTE LAYER

AGENTIC SWARM

  • > Fact-Discovery Agents (LangGraph)
  • > Contradiction & Propaganda Filters
  • > Ontology Engine (Actors, Drivers)
  • > Strategy / Policy Agents
OUTPUT LAYER

STRATEGIC ARTIFACTS

  • > Situation Reports (JSON / PDF)
  • > Stakeholder Maps
  • > Conflict Graphs
  • > Recommendations Engine
02. INGESTION VECTORS

>> Data Infrastructure

Tacitus uses three major pipelines, each tailored to a different temporal and epistemic profile:

Pipeline A: Velocity

TARGET: High-Frequency OSINT

Tavily / Serper / GDELT / AP Wire feed a real-time event stream. A deduplication engine filters noise every 3–12 hours, using text-similarity hashing and temporal clustering.

Example Sources: Reuters, AP, Al Jazeera, Anadolu, BBC Monitoring.

Pipeline B: Specificity

TARGET: Local Media / Niche Actors

Firecrawl extracts content from community radio, local newspapers, rebel statements, municipality bulletins, and NGO field reports.

Use Case: Enab Baladi (Syria), Congo Research Group, Sudanese neighborhood committees.

Pipeline C: Authority

TARGET: Slow-Moving High Authority RAG

A vector store of UN SG reports, ICCT research, Carnegie/ICG analyses, academic papers, legal frameworks. Indexed with Supabase or Vertex AI.

>> Pipeline Orchestration

All three pipelines flow through a LangGraph Orchestrator, which handles stateful multi-step reasoning, and a Cloud Run microservice that schedules asynchronous scraping tasks.

03. AGENTIC PROCESSING

>> The "Concordia" Swarm

Tacitus does not summarize — it reconstructs reality through a structured, multi-agent council. Agents cooperate, challenge each other, and escalate contradictions.

LAYER 1 — FACT DISCOVERY

Developments Agent

Synthesizes last 72h of political, military, and humanitarian movements.

Contradiction Checker

Runs cross-source validation via a ReAct chain. Flags propaganda, errors, or manipulated media.

LAYER 2 — DEEP ONTOLOGY

Conflict Ontology Engine

Converts events into an actor graph: state actors, militias, ministries, NGOs, business actors, and individual influencers.

Narrative Miner

Extracts legitimacy claims, grievances, power deltas, red lines. Distinguishes Drivers of Conflict from surface rhetoric.

LAYER 3 — POLICY GENERATION

> INPUT: User defines Stakeholder (UN, FCDO, NGO, CEO).

> PROCESS: Strategy Agent adjusts analysis to mandate, leverage, and risk tolerance.

> OUTPUT: Prioritized to-dos, mediation pathways, pressure levers, and anticipated reactions.

04. CONFLICT VISUALIZER

>> Visualizing Conflicts Across Scales

The Concordia Engine converts any conflict — geopolitical, organizational, familial — into a dynamic Actor → Interest → Constraint → Relationship graph. This visualization breaks down hidden structures, power asymmetries, and opportunities for mediation.

>> Node Details

Hover a node to see its Interests, Constraints, and Role.

>> Deploy the Engine

The Concordia Engine is available for institutional pilots. Supports JSON export, graph visualization, and multi-agent workflows.

Request Access