WormGuard & OpenClaw-Shield & Webgate (APNet 2026)

This repository provides the experimental code for poster papers accepted by APNet 2026:

1. WormGuard: Epidemic Modelling and Decentralized Containment of PromptWorms in AI Agent Networks

2. OpenClaw-Shield: Network-Layer Defence Against PromptWorm Propagation in Data Center Agent Deployments

3. WebGate: Browser-Native Edge Gateway for Local and Remote LLMs

---

Installation

The following software and libraries are required:

• Python 3.11
• PyTorch 2.5.1

Repository Structure

This repository contains the following example scripts:

• WormGuard.py
  Experimental code for the paper "WormGuard: Epidemic Modelling and Decentralized Containment of PromptWorms in AI Agent Networks".

• OpenClaw.py
  Experimental code for the paper "OpenClaw-Shield: Network-Layer Defence Against PromptWorm Propagation in Data Center Agent Deployments".

Usage

Running Python Scripts

Python scripts (e.g., WormGuard.py) can be executed in two ways.

1. Direct Execution

Open the file in a Python IDE such as PyCharm and run it directly.

2. Command-Line Execution

From a terminal, navigate to the script directory and run:

python WormGuard.py

The details of Webgate please check WebGate/Readme.