🧠 Cognitive User Simulation

An advanced LLM-powered conversational simulation framework for Discord

Featuring autonomous action planning, safety filtering, and multi-provider LLM integration for research and controlled testing

Features • Installation • Configuration • Usage • Documentation

DISCORD TERMS OF SERVICE VIOLATION RISK

This project is a SOCIAL EXPERIMENT and EDUCATIONAL TOOL designed to study AI interaction patterns.

BY USING THIS SOFTWARE, YOU ACKNOWLEDGE THAT:

1. ✅ You are using this strictly in private, controlled test environments
2. ❌ You will NOT use this on your main personal account (Risk of Permanent Ban)
3. ❌ You will NOT use this to spam, harass, or disrupt public servers
4. ⚖️ The author assumes NO RESPONSIBILITY for account bans, suspensions, or legal consequences

🔴 USE AT YOUR OWN RISK 🔴

• Python 3.8 or higher
• pip package manager
• A Discord Bot token

# 1. Clone the repository
git clone https://github.com/HighMark-31/Cognitive-User-Simulation.git
cd Cognitive-User-Simulation

# 2. Install dependencies
pip install -r requirements.txt

# 3. Configure environment
cp .env.example .env
# Edit .env with your credentials

# 4. Run the bot
python discord_bot.py

Create a .env file in the project root:

# ==========================================
# DISCORD CONFIGURATION
# ==========================================
DISCORD_BOT_TOKEN=your_bot_token_here

# ==========================================
# DATABASE CONFIGURATION
# ==========================================
DATABASE_URL=sqlite+aiosqlite:///discord_bot.db

# ==========================================
# LLM PROVIDER SELECTION
# ==========================================
# Options: glm | openai | gemini
LLM_PROVIDER=glm

# ==========================================
# GLM (Zhipu AI) CONFIGURATION
# ==========================================
GLM_API_KEY=your_glm_api_key_here
GLM_BASE_URL=https://z.ai/api/coding/paas/v4
GLM_MODEL=glm-4.7

# ==========================================
# OPENAI CONFIGURATION
# ==========================================
OPENAI_API_KEY=your_openai_api_key_here
OPENAI_BASE_URL=https://api.openai.com/v1
OPENAI_MODEL=gpt-5

# ==========================================
# GOOGLE GEMINI CONFIGURATION
# ==========================================
GEMINI_API_KEY=your_gemini_api_key_here
GEMINI_MODEL=gemini-3.0-flash

# ==========================================
# OPTIONAL SETTINGS
# ==========================================
PRIORITY_GUILD_ID=1320998163615846420

python discord_bot.py

The bot will automatically:

• 🔗 Connect to all servers it's present in
• 💬 Monitor messages in channels and DMs
• 🤖 Generate contextual responses using the configured LLM
• 💾 Log all interactions to the database

python stats_viewer.py

The bot autonomously decides what to do next using LLM-based planning:

When a user questions if the bot is an AI:

1. Automatically detects suspicion patterns
2. Logs the interaction to database
3. Generates a natural denial response for EXPERIMENTAL ONLY REASON
4. Saves data for analysis

Switch between providers by changing LLM_PROVIDER in .env:

LLM_PROVIDER=glm     # Z.ai GLM-4.7
LLM_PROVIDER=openai  # GPT-4o-mini
LLM_PROVIDER=gemini  # Gemini 1.5 Flash

Hard-coded regex patterns block:

• Discord invites
• PII (Personal Identifiable Information)
• Scam keywords
• Slurs and toxic content

• Typing Indicators: Simulates realistic typing time based on message length
• Reading Delays: Adds natural reading time before responding
• Presence Management: Rotates status (Online, Idle, DND) and activities
• Sleep Mode: Implements circadian rhythm with nighttime inactivity

┌─────────────────────────────────────┐
│ users                               │
├─────────────────────────────────────┤
│ Information about interacted users  │
└─────────────────────────────────────┘

┌─────────────────────────────────────┐
│ messages                            │
├─────────────────────────────────────┤
│ All received messages               │
└─────────────────────────────────────┘

┌─────────────────────────────────────┐
│ bot_responses                       │
├─────────────────────────────────────┤
│ All bot responses                   │
└─────────────────────────────────────┘

┌─────────────────────────────────────┐
│ interactions                        │
├─────────────────────────────────────┤
│ Interaction logs                    │
└─────────────────────────────────────┘

┌─────────────────────────────────────┐
│ social_test_logs                    │
├─────────────────────────────────────┤
│ General social test logs            │
└─────────────────────────────────────┘

All data is saved in discord_bot.db (SQLite).

Modify the bot's personality in discord_bot.py:

mood = 'chill'  # Options: chill, energetic, serious, playful
personality = 'sarcastic, direct, ironic, sometimes chaotic'

• 2-3 natural, colloquial sentences
• Reference concrete details from context
• Moderate use of slang/abbreviations
• Minimal emojis
• Adapt to channel language (EN/IT)

Examples:

• Verify the token in .env
• Check Discord Developer Portal privileges
• Review logs for errors
• Ensure bot has permission to read/send messages

• Verify the API key is valid
• Check available credits on your LLM account
• Verify the base URL is correct
• Test API connection manually

• Verify write permissions in the directory
• Check .env configuration
• Ensure SQLite is properly installed
• Review database.py for errors

1. Update LLM_PROVIDER in .env
2. Ensure the corresponding API key is set
3. Restart the bot
4. The bot will automatically initialize with the new provider

Cognitive-User-Simulation/
│
├── .env.example              # Example environment variables
│
├── discord_bot.py            # Main bot entry point
├── LLM_Client.py             # LLM provider integration
├── database.py               # Database models and operations
├── safety_filter.py          # Safety and content filtering
├── stats_viewer.py           # Statistics viewer utility
│
├── requirements.txt          # Python dependencies
├── LICENSE                   # Project license
└── README.md                 # This file

This project is for educational and testing purposes only. Use it responsibly.

See LICENSE file for more details.

For issues or questions:

• Check the logs in the database or debug.txt
• Open an issue in the repository
• Contact the author through GitHub