A collection of Python scripts for Discord data collection, trust analysis, and social ranking.

python3 -m venv venv
source ./venv/bin/activate

pip install -r requirements.txt

Dependencies:

• discord.py - Discord API wrapper
• python-dotenv - Environment variable management
• toml - Configuration file parsing
• pandas - Data processing
• scipy - Scientific computing (for score transformations)
• openai - OpenAI API (for post summarization)
• psycopg2 - PostgreSQL adapter (for database operations)

cp .env.example .env
# Edit .env and add your tokens

Required environment variables:

• DISCORD_TOKEN - Your Discord bot token
• DISCORD_CLIENT_ID - Your Discord application client ID (optional)
• DATABASE_URL - PostgreSQL connection string (for database scripts)
• OPENAI_API_KEY - OpenAI API key (for summarization)

[settings]
output_folder = "raw"
default_days = 730
max_messages_per_channel = 20000
channel_delay = 1
server_delay = 5

[exclusions]
exclude_channel_ids = [
    "channel_id_to_exclude",
]

[trust]
mention_points = 50
reply_points = 40
reaction_points = 30
here_points = 10
everyone_points = 10
role_points = 10

[servers]
server_ids = [
    "1021933844481445989",  # Your server ID
    "another_server_id",    # Add more servers
]

# Verify which channels the bot can access
python check_channels.py

# List all moderators and admins in servers
python find_mods.py
python find_mods.py --verbose  # Show detailed role information

# Process all servers from config.toml
python read_messages.py

# Override settings via command line
python read_messages.py --days 30 --max-messages 1000

# Calculate trust from collected messages
python generate_trust.py

# Use custom input path
python generate_trust.py --input raw

# Load collected JSON data into PostgreSQL
python load_discord_data.py

# Process scores with log transformation (default)
python process_scores.py

# Use square root transformation
python process_scores.py --sqrt

# Use quantile transformation
python process_scores.py --quantile

# Process seed CSV with default tier weights (60%, 20%, 20%)
python process_seed.py seed/my_seed.csv

# Custom tier weights
python process_seed.py seed/my_seed.csv --weights 0.5 0.3 0.2

# Generate JSON files for the UI
python generate_json.py

# Generate AI summaries for channel posts
python summarize_posts.py

socialrank/
├── raw/                    # Raw Discord message data (JSON)
├── trust/                  # Trust score CSVs per server
├── scores/                 # Score files for processing
├── output/                 # Processed score output
├── seed/                   # Seed data CSVs
├── ui/                     # Generated JSON for UI
├── schemas/                # SQL schema definitions
├── migrations/             # Database migrations
├── config.toml             # Main configuration file
├── requirements.txt        # Python dependencies
└── *.py                    # Python scripts

Messages saved as ServerName.json with structure:

{
  "server_info": { "id": "...", "name": "..." },
  "channels": {
    "channel_id": {
      "channel_info": { "name": "..." },
      "messages": [...]
    }
  }
}

User ID mappings saved as user_ids_ServerName.csv.

Trust scores saved as CSV files per server (e.g., karma3_labs.csv):

i,j,v
user_id_1,user_id_2,150
user_id_1,user_id_3,80

Where:

• i = user ID giving trust
• j = user ID receiving trust
• v = trust value (sum of all interactions)

Transformed and normalized scores with usernames:

i,v
username1,1000.00
username2,850.50

Combined seed and score data for the UI:

{
  "category": "socialrank",
  "server": "server_id",
  "seed": [{"i": "username", "v": 0.6}],
  "scores": [{"i": "username", "v": 950.0}]
}

Note: Bots are excluded from giving or receiving trust values.

The project includes PostgreSQL migrations in migrations/ and schemas in schemas/ for storing:

• Servers, channels, and users
• Messages with attachments, embeds, and reactions
• Mentions and role mentions
• Computed scores
• Channel summaries (AI-generated)

If the bot can't read messages:

1. Run python check_channels.py to see access status
2. Ensure the bot has "View Channel" and "Read Message History" permissions
3. Enable required intents in Discord Developer Portal

For find_mods.py, enable "Server Members Intent" at: https://discord.com/developers/applications/

Ensure DATABASE_URL is set correctly:

postgresql://user:password@host:port/database