Skip to content

configuration.md

Latest commit

 

History

History
105 lines (75 loc) · 5.18 KB

configuration.md

File metadata and controls

105 lines (75 loc) · 5.18 KB

Configuration Guide

This document describes how to configure Mirrobot for your Discord server.

Initial Setup

  1. Create a Discord bot application from the Discord Developer Portal
  2. Copy your bot token and add it to the .env file or directly into config.json
  3. Invite the bot to your server using the generated OAuth2 URL

Configuration File

The main configuration file is config.json. You can copy config_example.json as a starting point. Here's what each field means:

{
  "token": "See environment variable DISCORD_BOT_TOKEN",  // Your Discord bot token (RECOMMENDED: Use environment variable)
  "command_prefix": "!",                                  // Default command prefix
  "ocr_worker_count": 2,                                  // Number of parallel OCR workers
  "ocr_max_queue_size": 100,                              // Maximum OCR queue size
  "thread_purge_interval_minutes": 10,                    // How often to purge old thread data
  "ocr_read_channels": {},                                // Channels where OCR scanning is active
  "ocr_response_channels": {},                            // Channels where OCR responses are sent
  "ocr_response_fallback": {},                            // Fallback channels for OCR responses
  "server_prefixes": {},                                  // Server-specific command prefixes
  "command_permissions": {},                              // Command permission settings
  "ocr_channel_config": {},                               // OCR language settings per channel
  "maintenance": {                                        // Auto-restart and maintenance settings
    "restart_threshold_seconds": 86400,                   // Auto-restart after 24 hours
    "check_interval_minutes": 15,                         // Check interval for restart conditions
    "auto_restart_enabled": true                          // Enable automatic restarts
  }
}

Configuration Field Details

  • token: Your Discord bot token. STRONGLY RECOMMENDED to use the DISCORD_BOT_TOKEN environment variable instead
  • ocr_read_channels: Format: {"guild_id": [channel_id1, channel_id2]} - Channels to monitor for images
  • ocr_response_channels: Format: {"guild_id": [channel_id1, channel_id2]} - Where to send OCR results
  • ocr_response_fallback: Format: {"guild_id": [channel_id1, channel_id2]} - Backup channels if primary unavailable
  • server_prefixes: Format: {"guild_id": "prefix"} - Custom prefixes per server (e.g., "ocr!")
  • command_permissions: Format: {"guild_id": {"command": ["role_id1"]}} - Role-based command access
  • ocr_channel_config: Format: {"guild_id": {"channel_id": {"lang": "eng"}}} - Language per channel

Media Caching

The bot can be configured to re-upload Discord media attachments to third-party services to prevent links from expiring. This is useful for preserving conversation history that is passed to the LLM.

  • media_caching:
    • enabled: true or false. Enables or disables the media caching feature.
    • services: A list of services to use for uploading. Supported services are "litterbox", "catbox", and "pixeldrain".
    • upload_timeout_seconds: The timeout in seconds for each upload attempt.
    • permanent_host_fallback: true or false. If true, the bot will fall back to a permanent host if the temporary host (litterbox) fails.
    • pixeldrain_api_key: Your API key for the Pixeldrain service.
    • catbox_user_hash: Your user hash for the Catbox service.

Environment Variables

RECOMMENDED APPROACH: Use environment variables for sensitive data:

  1. Create a .env file based on the .env.example template
  2. Add your Discord bot token: DISCORD_BOT_TOKEN=your_token_here
  3. Optionally add Google AI API key: GOOGLE_AI_API_KEY=your_key_here

The bot will automatically use environment variables when available, overriding config file values.

OCR Configuration

Setting Up OCR Channels

To enable OCR in specific channels:

  1. Use !add_ocr_read_channel #channel-name to add a channel for OCR scanning
  2. Use !add_ocr_response_channel #channel-name to designate where responses go
  3. Optionally set language: !set_ocr_language #channel-name eng (eng = English, rus = Russian)

OCR Performance Tuning

  • Increase ocr_worker_count for faster processing (but higher CPU usage)
  • Adjust ocr_max_queue_size based on your server's activity level

Patterns Configuration

Patterns are stored in patterns.json. Each pattern associates a regex pattern with a response.

You can manage patterns using:

  • !add_response "response text" "name" "note"
  • !add_pattern_to_response "response-name" "regex pattern" "IGNORECASE|DOTALL"
  • !remove_pattern_from_response "response-name" pattern_id

Permissions

Manage command permissions:

  • !add_command_role @role command_name - Give a role permission to use a command
  • !remove_command_role @role command_name - Remove permission
  • !add_bot_manager @role - Give a role access to all non-system commands

Server-Specific Settings

  • Change the command prefix: !set_prefix $
  • Reset to default prefix: !reset_prefix
  • View all server settings: !server_info