• Python 3.8 or higher
• Git
• FFmpeg (Required for music features)
• Docker & Docker Compose (Optional, Recommended)

1. Clone the repository:

   git clone https://github.com/coffin399/llmcord-JP-plana
   cd llmcord-JP-plana

2. Configure config.yaml:

   Copy config.default.yaml to create config.yaml.

   cp config.default.yaml config.yaml

   Open the generated config.yaml and configure at least the following settings:

   • bot_token: Required. Discord Bot Token
   • llm: section: model, providers (API keys, etc.)

🚀 Windows (Easy)

# Double-click startPLANA.bat

💻 Standard Method

pip install -r requirements.txt
python main.py

🐳 Docker (Recommended)

docker compose up --build -d

For additional features (Twitch notifications, Media downloader, TTS), please refer to the language-specific documentation.

• Multiple AI Models: OpenAI GPT-4o, Google Gemini 2.5 Pro, NVIDIA NIM, Mistral, and more
• Image Recognition: AI-powered image understanding and analysis
• Image Generation: High-quality image generation using Stable Diffusion WebUI Forge
• Web Search: Real-time internet search capabilities
• Memory System: User bio, channel bio, and global memory management
• Multi-language Support: Automatically adapts to user's language

• Multiple Sources: YouTube, SoundCloud, Niconico, and other platforms
• Advanced Controls: Play, pause, skip, volume, seek, loop modes
• Queue Management: Shuffle, remove, clear, and playlist support
• Auto Management: Automatic voice channel management

• Valorant: Player statistics tracking and display
• Rainbow Six Siege: Game statistics and performance metrics

• Style-Bert-VITS2: High-quality Japanese voice synthesis
• Voice Notifications: Join/leave announcements in voice channels
• Customizable: Speech rate, style, and emotion adjustments

• Earthquake Alerts: Japan earthquake and tsunami notifications with maps
• Twitch Streams: Automatic stream start notifications

• Akinator: Character guessing game with multi-language support
• Gacha System: Blue Archive-style student recruitment simulation
• Image Commands: Cat images, anime image search, and more

• Server Management: Server info, user info, avatar display
• Dice Rolling: Advanced dice roll system with custom notation
• Timer: Countdown timer functionality
• Media Downloader: Download videos and audio from various platforms

• Framework: discord.py 2.0+
• Language: Python 3.8+
• Configuration: YAML-based settings
• Logging: Discord channel logging with rate limiting
• Error Handling: Comprehensive exception handling system

• OpenAI: GPT-4o, GPT-4 Turbo
• Google: Gemini 2.5 Pro/Flash
• NVIDIA NIM: Various open-source models
• Mistral: Mistral Medium
• HenrikDev: Valorant API
• TheCatAPI: Cat image API
• P2P Earthquake: Japan earthquake data

llmcord-JP-plana/
├── main.py                    # Main entry point
├── config.default.yaml        # Default configuration
├── config.yaml               # Actual configuration (create this)
├── requirements.txt          # Python dependencies
├── startPLANA.bat           # Windows startup script
├── PLANA/                   # Main feature modules
│   ├── llm/                 # AI chat functionality
│   ├── music/               # Music playback
│   ├── images/              # Image commands
│   ├── notifications/       # Notification features
│   ├── tracker/             # Game trackers
│   ├── tts/                 # Text-to-speech
│   ├── utilities/           # Utility commands
│   └── services/            # Common services
├── modules/                 # Legacy modules
├── plugins/                 # Plugins
├── data/                    # Data storage
└── cache/                   # Cache directory

PLANA requires the Message Content Intent for the following features:

• Messages that @mention PLANA
• Replies to PLANA's messages
• Messages in channels where text-to-speech is enabled

No other server messages are collected or stored.

• Discord: coffin299
• X (Twitter): @coffin299
• GitHub Issues: Report Issues
• In-Bot Support: Use /support command

This project is released under the MIT License.

Developed by ごみぃ (coffin299) & えんじょ (Autmn134F)

Based on llmcord