---
title: README
emoji: 🧬
colorFrom: gray
colorTo: green
sdk: static
pinned: false
---

# 🧠 CyberDeathSec

> *We automate resistance. We build offensive AI. We open-source cyber warfare tools.*

---

## ⚙️ Mission

CyberDeathSec exists to weaponize machine learning for cybersecurity.  
We fine-tune large language models for:

- Red teaming & pentesting
- OSINT & footprinting
- Threat intelligence & CVE parsing
- Automation of exploit chains and tooling

---

## 📁 Projects

| Model | Description | Status |
|-------|-------------|--------|
| [`pentest-agent`](https://huggingface.co/7h3-R3v3n4n7/pentest-agent) | LLaMA 3-based model for offensive security and vulnerability triage | 🛠 In Progress |
| `osint-bot` | OSINT-focused LLM designed for profiling and recon | 🧪 Prototype |

---

## 🛠 Stack

```txt
🧠 Models:     LLaMA 3
⚙️  Tuning:     LoRA / PEFT / Unsloth
📦 Formats:    GGUF, Alpaca, ChatML
📊 Datasets:   CVE, NVD, STIX 2.1, ExploitDB
🚀 Deployment: LM Studio, Discord bots
```

---

## 🏴 Philosophy

```txt
We fight surveillance with intelligence.
We are CyberDeathSec.
We are Anonymous.
We are Legion.
We do not forgive.
We do not forget.
Expect us.
```

---

## 🤝 Join Us
```txt
Want to collaborate, contribute, or fork our work?

GitHub: CyberDeathSec

Models: Hugging Face

PRs welcome — we support the open tools arms race.
```

> 🕶 *Encrypted minds. Tactical AI. Ghosts in the machine.*