ghosthets commited on
Commit
8fc9f83
ยท
verified ยท
1 Parent(s): fb6d3ed

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +52 -33
README.md CHANGED
@@ -29,32 +29,38 @@ tags:
29
  - Dec-community
30
  - Dextron
31
  ---
32
- # ๐Ÿ›ก๏ธ Dex โ€” Your Personal Cybersecurity AI Assistant
33
 
34
- Dex (Digital Exploit eXpert) is a conversational AI assistant designed for cybersecurity enthusiasts, CTF players, bug bounty hunters, and students. Trained to assist with technical queries, it can chat like a pro while helping with security topics.
 
 
35
 
36
  ---
37
 
38
- ## ๐Ÿ“Š Model Info
39
 
40
- - **Model Name:** ghosthets/Dex
41
- - **Type:** Causal Language Model
42
- - **Skills:** General conversation, cybersecurity explanations, basic exploit reasoning, CTF hints, code logic
43
- - **Trained for:** Educational and conversational use only
 
 
 
44
 
45
  ---
46
 
47
- ## ๐Ÿง  Dex Can Help With
48
 
49
- - โœ… Cybersecurity concepts (XSS, CSRF, SQLi, etc.)
50
- - โœ… CTF challenges & walkthroughs
51
- - โœ… Payload generation tips
52
- - โœ… Code logic explanations
53
- - โœ… Creating custom chatbot assistants
 
54
 
55
  ---
56
 
57
- ## โš™๏ธ Usage with Transformers
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
- decoded = tokenizer.decode(output[0], skip_special_tokens=True)
85
- reply = decoded.split("Dex:")[-1].strip()
86
- return reply
87
-
88
- print(ask_dex("Explain SQL Injection in simple terms"))
89
  ```
90
 
91
- #๐Ÿ‘จโ€๐Ÿ’ป Made For
92
- Ethical Hackers
 
 
 
93
 
94
- Cybersecurity Learners
95
 
96
- Tech Students
 
 
 
97
 
98
- Curious Minds
 
 
 
 
 
 
 
 
 
 
 
 
 
99
 
100
- โšก Optimized for edge devices โ€” This model is fine-tuned to run efficiently even on low-power systems like Raspberry Pi Zero W, 2W, and similar micro boards. While it's currently trained on limited data, we plan to expand its training set soon, making it even more accurate, interactive, and capable of deeper conversations and secure communications.
101
 
102
- #๐Ÿง  Creator
103
- Gaurav Chouhan
104
- aka ghosthets
105
- ๐Ÿ”— GitHub: ghosthets
106
- ๐Ÿ“ India ๐Ÿ‡ฎ๐Ÿ‡ณ
 
29
  - Dec-community
30
  - Dextron
31
  ---
32
+ # ๐Ÿ›ก๏ธ Dex โ€” Your Personal Cybersecurity AI Sidekick
33
 
34
+ > *โ€œBuilt 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
+ ## ๐Ÿง  Model Intelligence
41
 
42
+ | Feature | Description |
43
+ | --------- | ------------------------------------------------------- |
44
+ | ๐Ÿ”Ž Name | `ghosthets/Dex` |
45
+ | ๐Ÿง  Type | Causal Language Model |
46
+ | ๐ŸŽฏ Domain | Cybersecurity, CTFs, Ethical Hacking |
47
+ | ๐Ÿงช Usage | Educational, research, AI assistant for cyber learners |
48
+ | ๐Ÿงฐ Tools | Text reasoning, payload discussion, exploit explanation |
49
 
50
  ---
51
 
52
+ ## ๐Ÿ’ฅ Dex Is Trained To:
53
 
54
+ * ๐Ÿ’ฌ Talk like a pro on cybersecurity topics
55
+ * ๐Ÿ” Explain CVEs, XSS, CSRF, SQLi, etc.
56
+ * ๐Ÿงน Assist in CTF logic & recon mindset
57
+ * ๐Ÿ’ก Generate payload hints
58
+ * ๐Ÿ‘พ Help you build your own AI security bots
59
+ * โš™๏ธ Run locally with just CPU or on edge boards
60
 
61
  ---
62
 
63
+ ## โš™๏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
+ ## โšก Edge-Device Ready
95
+
96
+ > ๐Ÿ’ป **Lightweight. Efficient. Ready-to-Hack.**
97
 
98
+ Dex is **engineered to run smoothly on low-power devices** like:
99
 
100
+ * ๐Ÿ”น Raspberry Pi Zero W
101
+ * ๐Ÿ”น Pi 2W & Pi 4
102
+ * ๐Ÿ”น Tinker Boards / NanoPi / Pocket PCs
103
+ * ๐Ÿ”น 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
+ ## ๐Ÿ‘จโ€๐Ÿ’ป 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
+ ## ๐Ÿง  Built With โค๏ธ by
121
 
122
+ **Gaurav Chouhan**
123
+ aka `ghosthets`
124
+ ๐Ÿ”— [GitHub: ghosthets](https://github.com/ghosthets)
125
+ ๐Ÿ“ From India ๐Ÿ‡ฎ๐Ÿ‡ณ