Engine Layer

The Engine Layer powers the core logic of the Tythe OS responsible for transforming verified behavior into programmable credibility. It is a modular scoring system that ingests trust signals, assigns weight, and outputs a real-time DISC Score for every Individual and AI Agent TRIS ID.

The Engine Layer turns raw data into structured, zero-knowledge-backed assessments, enabling composable, reputation-aware logic across all downstream systems.

What the Engine Does

Tythe’s DISC Engine performs three primary functions:

1. Process Input Data

Input Type	Description
Verifications	ZK-proven identity or account claims (e.g., GitHub, X)
Trovebook	ZK-hashed submissions of contributions (Individuals only)
Decision Relays	Platform-submitted attestations of real-time actions or contributions

Note: Sponsorships are not input data. They function as a separate scoring vector for AI Agents (see below for more details).

2. Score Based on Logic and Weighting

Tythe applies two distinct scoring models:

Metric-Weighted Scoring (for Individuals + AI Agents):

• Inputs (Verifications, Trovebook, Relays) are:

  • Mapped to one or more of the 7 DISC metric domains

  • Weighted by criticality using platform-defined LogicCR

  • Converted into TCT → DISC Score

Sponsorship-Based Scoring (AI Agents only):

• AI Agents can be sponsored by verified Individuals or Organizations

• Sponsorships allocate TCT directly to the agent’s Credibility Vault

• Scoring is based on:

  • Sponsor type (Org vs Individual)

  • TCT amount

  • Sponsor DISC Score or Organizational Rank

Sponsorships are excluded from the metric-tagging model. They contribute to score via a dedicated volume- and trust-weighted formula.

3. Output and Enforce DISC Scores

• Scores are recorded onchain

• Updated in real time

• Queryable via:

  • DISC Reports (Individuals only)

  • Indexes, APIs, and dashboards (Individuals + AI Agents)

All Scoring Is:

• Modular — Inputs are mapped to specific metric categories

• Impact-Weighted — Actions are scored using LogicCR-defined criticality

• ZK-Powered — No raw data is exposed; only verifiable hashed proofs

• Tamper-Resistant — Deterministic logic prevents gaming or inflation

• Eligibility-Gated —

  • Individuals must complete zk-KYH to begin scoring

  • AI Agents must complete zk-KYA to begin scoring

• Dual-Path Scoring —

  • Individuals earn via input-weighted scoring (Trovebook, Verifications, Relays)

  • AI Agents earn via Decision Relays and TCT-backed Sponsorships

Engine Layer Components

Component	Description
Scoring Logic	The operational brain of the DISC Engine. Divided into Pre-Protocol (for legacy contributions) and Post-Protocol (for real-time actions).
Scoring Metrics	The seven modular trust categories that define how credibility is scored: Financial, Creative, Educational, Behavioral, Social, Security, and Compliance.

How It Fits in the OS Stack

The Engine Layer sits between:

• Input Layer

  → Accepts Verifications, Trovebook entries, and Decision Relays

• Output Layer

  → Produces:

  • DISC Scores for Individuals and AI Agents

  • DISC Reports (for Individuals only)

  • Queryable outputs for Indexes, platforms, and GateCR logic

This is the stage where credibility becomes code; structured, weighted, and enforced as a composable protocol primitive.

---

“The Lord detests differing weights, and dishonest scales do not please him.”

— Proverbs 20:23