Worm-AI is a command-line interface (CLI) for interacting with Grok models, utilizing an unofficial, reverse-engineered API wrapper. It targets users seeking a lightweight, terminal-native experience for LLM interaction, offering features like built-in jailbreak capabilities and a customizable user interface for fast and flexible engagement.

How It Works

The core of Worm-AI is its reliance on a reverse-engineered Grok API wrapper, originally developed by realasfngl/Grok-Api. This wrapper acts as a proxy, enabling the CLI to access Grok endpoints without requiring official API credentials. This design choice facilitates a lightweight, powerful, and terminal-native interaction model, prioritizing speed and flexibility.

Quick Start & Requirements

• Installation: Clone the repository (git clone https://github.com/kafyasfngl/worm-ai), cd worm-ai, then pip install -r requirements.txt. Run the CLI via python3 main.py.
• Prerequisites: Python 3.8+ and pip.
• Dependencies: Specified in requirements.txt.
• Links: Project URL: https://github.com/kafyasfngl/worm-ai.

Highlighted Details

• Unofficial Grok API wrapper (reverse-engineered backend).
• Integrated jailbreak system for unrestricted responses.
• Fully customizable terminal UI, including banner, prompt style, and typing animations.
• Modular architecture allowing for easy replacement of backends or UI components.

Maintenance & Community

• Author: @kafyasfngl.
• Contributors: @neo.
• Community: Telegram group available at t.me/xsocietyforums.

Licensing & Compatibility

• License: No explicit license is declared in the provided README.
• Compatibility: Intended for research, educational, and local testing. Use is subject to API provider terms; avoid misuse.

Limitations & Caveats

This project is unofficial and not affiliated with xAI or Grok. It depends on an unofficial, reverse-engineered API, which may be unstable or subject to change without notice. Security recommendations include storing private configurations in .env files and avoiding public commits of sensitive data.