DETERMINISTIC & PHYSICS-FIRST RESEARCH ENGINES

Threat Intelligence Lab

Explore the algorithmic foundations of the Aegis Protocol: Mandelbrot token frequency regression, Hawkes point-process contagion physics, and Byzantine fault-tolerant swarm consensus.

// INSTRUMENT_01: STATISTICAL_FINGERPRINTING
Mandelbrot Power-Law Regression
POWER-LAW SCANNER ARMED

Autoregressive LLM transformers sample token sequences with strictly bounded power-law decay ($P(r) = P_0(r+\beta)^{-\gamma}$) and suppressed lexical entropy. Human composition exhibits spontaneous burstiness and colloquial dispersion.

QUICK LOAD DATASET SAMPLES:
Tokens: 0 | Unique: 0
// INSTRUMENT_02: STOCHASTIC_EPIDEMIOLOGY
Hawkes Process Contagion Simulator
POINT PROCESS READY

Simulates viral cascade dynamics via self-exciting Hawkes point processes: $\lambda(t) = \mu + \sum_{t_i < t} \alpha e^{-\beta(t - t_i)}$. Computes effective reproduction number $R_0 = \alpha / \beta$ and simulates live network node infection and autonomous defense swarm containment.

SELECT PRESET THREAT VECTOR:
// INSTRUMENT_03: BYZANTINE_FAULT_TOLERANCE
Byzantine Swarm Consensus Engine
W-MSR FILTER ACTIVE

Executes parallel adversarial verification across multi-agent nodes. Uses Weighted Mean Subsequence Reduction (W-MSR) to prune compromised, hallucinating, or sycophantic outlier nodes before calculating consensus.

TEST CLAIMS: