VAC Live + Overwatch 2.0: Hybrid Anti-Cheat System (Proposal)

Author: (steam mishka.sit2002) Version: 3.5 (Final)
Steam: Mishka.sit2002

---

🚀 Overview

This repository contains a technical proposal and a Python-based proof-of-concept for a hybrid anti-cheat system designed for Counter-Strike 2. The goal is to close the "arms race" gap where cheat developers adapt faster than machine learning (VAC Live) can retrain.

🧠 Key Features

1. Glicko-2 Judge Rating

Unlike the old Overwatch, this system uses the Glicko-2 algorithm to rate judges based on their accuracy.

• Bot-Farm Protection: High entry barriers (wins/hours) and weighted voting make bribery or botting economically impossible.
• Vote Weight: A Senior Judge's verdict carries more weight than a Trainee's.

2. Invisible Honeypot Entities (Deterministic Detection)

A method to provide mathematical proof of cheating.

• Invisible "honeypot" entities are placed inside geometry or behind walls.
• If a player's telemetry shows consistent snapping or tracking of these invisible targets, the system records a 100% deterministic violation.

3. Shadow Monitoring (2-3 Match Verification)

To eliminate False Positives:

• If found "Guilty" by consensus, the account is shadow-flagged.
• It remains under surveillance for the next 2-3 matches.
• The ban is only finalized if cheating patterns (AIM/WH anomalies) persist.

---

🛠 Movement Bug Fix (Bonus)

This proposal also includes a logic fix for the "Fall Damage" bug in Source 2:

• The Problem: Engine calculates damage based on "air-time," leading to deaths while sliding on boxes.
• The Fix: Switch to a Vertical Height Delta (Z-delta) method. Reset the fall interval on any solid contact (ledge, slope, or box edge).

---

💻 How to use this repository

1. Read the documentation above for the logic.
2. Run prototype.py (to be added) to see the Python simulation of judge ratings and case processing.