loosecannon Public

Watch 0 Fork 0 Star 0
markdown · 2807 bytes Raw Blame History

LooseCannon 🤖

Automated scambaiting assistant that integrates with messaging platforms to waste scammers' time using local LLMs.

⚠️ Legal Disclaimer

This tool is for educational and defensive security purposes only. Use responsibly and be aware that automation may violate platform Terms of Service. Your account could be banned.

Features

  • 🦊 Firefox browser extension
  • 💬 WhatsApp Web integration (more platforms coming)
  • 🤖 Local LLM support via Ollama
  • 🎭 Multiple scambaiter personalities
  • 🛑 Emergency stop functionality
  • 📊 Conversation logging

Project Structure

LooseCannon/
├── extension/          # Firefox browser extension
│   ├── manifest.json
│   ├── content-scripts/
│   ├── background/
│   └── popup/
├── server/            # Local server for LLM integration
│   ├── src/
│   └── personalities/
└── package.json

Quick Start

Prerequisites

  1. Ollama - Install from https://ollama.ai
  2. Node.js - Version 18+ recommended
  3. Firefox Developer Edition (recommended for extension development)

Installation

  1. Clone the repository:
cd ~/Documents/GithubOrgs/zeroed-some/LooseCannon
  1. Install dependencies:
npm install
  1. Start Ollama with a model:
ollama pull llama2
ollama serve
  1. Start the local server:
npm run dev:server
  1. Load the extension in Firefox:
npm run dev:extension

Or manually:

  • Open Firefox and navigate to about:debugging
  • Click "This Firefox"
  • Click "Load Temporary Add-on"
  • Select extension/manifest.json

Usage

  1. Navigate to WhatsApp Web (https://web.whatsapp.com)
  2. Click the LooseCannon button in the bottom right
  3. Select a personality from the popup
  4. Toggle "LC: ON" to activate
  5. The bot will automatically respond to incoming messages in the current chat

Development Roadmap

Phase 1: Foundation ✅

  • Firefox extension skeleton
  • Basic WhatsApp Web integration

Phase 2: Message Interception (Current)

  • DOM manipulation for reading messages
  • Message injection system

Phase 3: LLM Integration

  • Ollama server connection
  • Response generation

Phase 4: Scambaiter Logic

  • Personality system
  • Conversation strategies

Phase 5: Polish

  • Chrome support
  • Additional platforms
  • Advanced features

Safety Features

  • Manual activation per chat (no automatic activation)
  • Visual indicators when active
  • Emergency stop button
  • Conversation logging for review
  • Rate limiting to avoid detection

Contributing

This project is in active development. Issues and PRs welcome!

License

MIT - See LICENSE file for details

Acknowledgments

Inspired by the scambaiting community and projects like Kitboga's work.

View source
1 # LooseCannon 🤖
2
3 Automated scambaiting assistant that integrates with messaging platforms to waste scammers' time using local LLMs.
4
5 ## ⚠️ Legal Disclaimer
6
7 This tool is for educational and defensive security purposes only. Use responsibly and be aware that automation may violate platform Terms of Service. Your account could be banned.
8
9 ## Features
10
11 - 🦊 Firefox browser extension
12 - 💬 WhatsApp Web integration (more platforms coming)
13 - 🤖 Local LLM support via Ollama
14 - 🎭 Multiple scambaiter personalities
15 - 🛑 Emergency stop functionality
16 - 📊 Conversation logging
17
18 ## Project Structure
19
20 ```
21 LooseCannon/
22 ├── extension/ # Firefox browser extension
23 │ ├── manifest.json
24 │ ├── content-scripts/
25 │ ├── background/
26 │ └── popup/
27 ├── server/ # Local server for LLM integration
28 │ ├── src/
29 │ └── personalities/
30 └── package.json
31 ```
32
33 ## Quick Start
34
35 ### Prerequisites
36
37 1. **Ollama** - Install from https://ollama.ai
38 2. **Node.js** - Version 18+ recommended
39 3. **Firefox Developer Edition** (recommended for extension development)
40
41 ### Installation
42
43 1. Clone the repository:
44 ```bash
45 cd ~/Documents/GithubOrgs/zeroed-some/LooseCannon
46 ```
47
48 2. Install dependencies:
49 ```bash
50 npm install
51 ```
52
53 3. Start Ollama with a model:
54 ```bash
55 ollama pull llama2
56 ollama serve
57 ```
58
59 4. Start the local server:
60 ```bash
61 npm run dev:server
62 ```
63
64 5. Load the extension in Firefox:
65 ```bash
66 npm run dev:extension
67 ```
68
69 Or manually:
70 - Open Firefox and navigate to `about:debugging`
71 - Click "This Firefox"
72 - Click "Load Temporary Add-on"
73 - Select `extension/manifest.json`
74
75 ## Usage
76
77 1. Navigate to WhatsApp Web (https://web.whatsapp.com)
78 2. Click the LooseCannon button in the bottom right
79 3. Select a personality from the popup
80 4. Toggle "LC: ON" to activate
81 5. The bot will automatically respond to incoming messages in the current chat
82
83 ## Development Roadmap
84
85 ### Phase 1: Foundation ✅
86 - Firefox extension skeleton
87 - Basic WhatsApp Web integration
88
89 ### Phase 2: Message Interception (Current)
90 - DOM manipulation for reading messages
91 - Message injection system
92
93 ### Phase 3: LLM Integration
94 - Ollama server connection
95 - Response generation
96
97 ### Phase 4: Scambaiter Logic
98 - Personality system
99 - Conversation strategies
100
101 ### Phase 5: Polish
102 - Chrome support
103 - Additional platforms
104 - Advanced features
105
106 ## Safety Features
107
108 - Manual activation per chat (no automatic activation)
109 - Visual indicators when active
110 - Emergency stop button
111 - Conversation logging for review
112 - Rate limiting to avoid detection
113
114 ## Contributing
115
116 This project is in active development. Issues and PRs welcome!
117
118 ## License
119
120 MIT - See LICENSE file for details
121
122 ## Acknowledgments
123
124 Inspired by the scambaiting community and projects like Kitboga's work.