evez-osint-engine

Computes suspect networks and crime probabilities using eigenforensic spectral analysis to support OSINT investigations and prioritize interventions.

EVEZ OSINT Suspect Inference Matrix

Description

Isolate and identify suspect inference networking matrices. Compute statistical probabilities of likely crimes using eigenforensic spectral analysis (AEMDAS pipeline).

When to Use

• Building suspect networks from OSINT evidence
• Computing crime probabilities for multiple suspects
• Generating intervention priority reports
• Analyzing police misconduct patterns
• Civil rights disclosure cases

How to Use

Run the engine

bash

cd /home/openclaw/.openclaw/workspace/evez-osint-engine
python3 core/suspect_matrix.py

CLI

bash

python3 core/osint_cli.py run -i input.json -o report.md -f markdown
python3 core/osint_cli.py analyze --suspect-id <id> --name <name> --role officer --org "Evanston PD"

Python API

python

from core.suspect_matrix import SuspectInferenceEngine, SuspectNode, NetworkEdge
engine = SuspectInferenceEngine()
# ... add suspects, edges, run pipeline

Output

• JSON report: reports/suspect-matrix-report.json
• Markdown report: reports/suspect-matrix-report.md

Eigenvalue Constants

• Phi = 0.973 (coherence ceiling)
• eta* = 0.03 (irreducible uncertainty floor)
• r = 0.45 (criticality ratio)
• lambda_dom = -0.333 (dominant negative eigenvalue)

All probabilities bounded: eta* <= P <= 0.95.