Update README.md
Browse files
README.md
CHANGED
|
@@ -29,32 +29,38 @@ tags:
|
|
| 29 |
- Dec-community
|
| 30 |
- Dextron
|
| 31 |
---
|
| 32 |
-
# ๐ก๏ธ Dex โ Your Personal Cybersecurity AI
|
| 33 |
|
| 34 |
-
|
|
|
|
|
|
|
| 35 |
|
| 36 |
---
|
| 37 |
|
| 38 |
-
##
|
| 39 |
|
| 40 |
-
|
| 41 |
-
|
| 42 |
-
|
| 43 |
-
|
|
|
|
|
|
|
|
|
|
| 44 |
|
| 45 |
---
|
| 46 |
|
| 47 |
-
##
|
| 48 |
|
| 49 |
-
|
| 50 |
-
|
| 51 |
-
|
| 52 |
-
|
| 53 |
-
|
|
|
|
| 54 |
|
| 55 |
---
|
| 56 |
|
| 57 |
-
## โ๏ธ
|
| 58 |
|
| 59 |
```python
|
| 60 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
@@ -62,15 +68,13 @@ import torch
|
|
| 62 |
|
| 63 |
tokenizer = AutoTokenizer.from_pretrained("ghosthets/Dex")
|
| 64 |
model = AutoModelForCausalLM.from_pretrained("ghosthets/Dex")
|
| 65 |
-
|
| 66 |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
| 67 |
model.to(device)
|
| 68 |
|
| 69 |
def ask_dex(prompt):
|
| 70 |
input_text = f"User: {prompt}\nDex:"
|
| 71 |
inputs = tokenizer(input_text, return_tensors="pt").to(device)
|
| 72 |
-
|
| 73 |
-
output = model.generate(
|
| 74 |
inputs.input_ids,
|
| 75 |
attention_mask=inputs.attention_mask,
|
| 76 |
max_length=256,
|
|
@@ -80,27 +84,42 @@ def ask_dex(prompt):
|
|
| 80 |
temperature=0.7,
|
| 81 |
pad_token_id=tokenizer.eos_token_id
|
| 82 |
)
|
|
|
|
| 83 |
|
| 84 |
-
|
| 85 |
-
reply = decoded.split("Dex:")[-1].strip()
|
| 86 |
-
return reply
|
| 87 |
-
|
| 88 |
-
print(ask_dex("Explain SQL Injection in simple terms"))
|
| 89 |
```
|
| 90 |
|
| 91 |
-
|
| 92 |
-
|
|
|
|
|
|
|
|
|
|
| 93 |
|
| 94 |
-
|
| 95 |
|
| 96 |
-
|
|
|
|
|
|
|
|
|
|
| 97 |
|
| 98 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 99 |
|
| 100 |
-
|
| 101 |
|
| 102 |
-
|
| 103 |
-
|
| 104 |
-
|
| 105 |
-
|
| 106 |
-
๐ India ๐ฎ๐ณ
|
|
|
|
| 29 |
- Dec-community
|
| 30 |
- Dextron
|
| 31 |
---
|
| 32 |
+
# ๐ก๏ธ Dex โ Your Personal Cybersecurity AI Sidekick
|
| 33 |
|
| 34 |
+
> *โBuilt for the curious, optimized for the underground.โ*
|
| 35 |
+
|
| 36 |
+
**Dex** (Digital Exploit eXpert) is an intelligent cybersecurity-oriented conversational AI built for ethical hackers, CTF warriors, and knowledge-hungry learners. Whether you're crafting payloads, understanding CVEs, or just chatting โ Dex makes cyber learning **fun, fast & fearless**.
|
| 37 |
|
| 38 |
---
|
| 39 |
|
| 40 |
+
## ๐ง Model Intelligence
|
| 41 |
|
| 42 |
+
| Feature | Description |
|
| 43 |
+
| --------- | ------------------------------------------------------- |
|
| 44 |
+
| ๐ Name | `ghosthets/Dex` |
|
| 45 |
+
| ๐ง Type | Causal Language Model |
|
| 46 |
+
| ๐ฏ Domain | Cybersecurity, CTFs, Ethical Hacking |
|
| 47 |
+
| ๐งช Usage | Educational, research, AI assistant for cyber learners |
|
| 48 |
+
| ๐งฐ Tools | Text reasoning, payload discussion, exploit explanation |
|
| 49 |
|
| 50 |
---
|
| 51 |
|
| 52 |
+
## ๐ฅ Dex Is Trained To:
|
| 53 |
|
| 54 |
+
* ๐ฌ Talk like a pro on cybersecurity topics
|
| 55 |
+
* ๐ Explain CVEs, XSS, CSRF, SQLi, etc.
|
| 56 |
+
* ๐งน Assist in CTF logic & recon mindset
|
| 57 |
+
* ๐ก Generate payload hints
|
| 58 |
+
* ๐พ Help you build your own AI security bots
|
| 59 |
+
* โ๏ธ Run locally with just CPU or on edge boards
|
| 60 |
|
| 61 |
---
|
| 62 |
|
| 63 |
+
## โ๏ธ Run It in Python (Transformers)
|
| 64 |
|
| 65 |
```python
|
| 66 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
|
|
|
| 68 |
|
| 69 |
tokenizer = AutoTokenizer.from_pretrained("ghosthets/Dex")
|
| 70 |
model = AutoModelForCausalLM.from_pretrained("ghosthets/Dex")
|
|
|
|
| 71 |
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
|
| 72 |
model.to(device)
|
| 73 |
|
| 74 |
def ask_dex(prompt):
|
| 75 |
input_text = f"User: {prompt}\nDex:"
|
| 76 |
inputs = tokenizer(input_text, return_tensors="pt").to(device)
|
| 77 |
+
outputs = model.generate(
|
|
|
|
| 78 |
inputs.input_ids,
|
| 79 |
attention_mask=inputs.attention_mask,
|
| 80 |
max_length=256,
|
|
|
|
| 84 |
temperature=0.7,
|
| 85 |
pad_token_id=tokenizer.eos_token_id
|
| 86 |
)
|
| 87 |
+
return tokenizer.decode(outputs[0], skip_special_tokens=True).split("Dex:")[-1].strip()
|
| 88 |
|
| 89 |
+
print(ask_dex("How to identify an XSS vulnerability?"))
|
|
|
|
|
|
|
|
|
|
|
|
|
| 90 |
```
|
| 91 |
|
| 92 |
+
---
|
| 93 |
+
|
| 94 |
+
## โก Edge-Device Ready
|
| 95 |
+
|
| 96 |
+
> ๐ป **Lightweight. Efficient. Ready-to-Hack.**
|
| 97 |
|
| 98 |
+
Dex is **engineered to run smoothly on low-power devices** like:
|
| 99 |
|
| 100 |
+
* ๐น Raspberry Pi Zero W
|
| 101 |
+
* ๐น Pi 2W & Pi 4
|
| 102 |
+
* ๐น Tinker Boards / NanoPi / Pocket PCs
|
| 103 |
+
* ๐น Light virtual containers and offline setups
|
| 104 |
|
| 105 |
+
Despite its small size, Dex delivers powerful outputs and will soon be trained on **more curated data**, making it **sharper, smarter, and more secure**.
|
| 106 |
+
|
| 107 |
+
---
|
| 108 |
+
|
| 109 |
+
## ๐จโ๐ป Perfect For
|
| 110 |
+
|
| 111 |
+
* Ethical Hackers
|
| 112 |
+
* Red Teamers & Pentesters
|
| 113 |
+
* Bug Bounty Hunters
|
| 114 |
+
* Cybersecurity Students
|
| 115 |
+
* AI + Security Researchers
|
| 116 |
+
* Solo Cyber Ninjas
|
| 117 |
+
|
| 118 |
+
---
|
| 119 |
|
| 120 |
+
## ๐ง Built With โค๏ธ by
|
| 121 |
|
| 122 |
+
**Gaurav Chouhan**
|
| 123 |
+
aka `ghosthets`
|
| 124 |
+
๐ [GitHub: ghosthets](https://github.com/ghosthets)
|
| 125 |
+
๐ From India ๐ฎ๐ณ
|
|
|