SilkroadBot

SilkroadBot - Automation Framework

⚔️ A free, open-source Silkroad Online automation framework built in C# for Windows. Inspired by modular architectures like RSBot, with an integrated AI decision engine powered by Google Gemini.

🏗️ Architecture

SilkroadBot/
├── src/
│   ├── SilkroadBot.Core          # Bot engine, networking, crypto, events, plugin hosting
│   ├── SilkroadBot.Domain        # Game models, enums, protocol adapters
│   ├── SilkroadBot.Plugins.SDK   # Plugin development SDK
│   ├── SilkroadBot.Navigation    # Pathfinding (A*, NavMesh-ready)
│   ├── SilkroadBot.AI            # AI integration (Gemini) + Guardrails
│   ├── SilkroadBot.Profiles      # Multi-profile management
│   └── SilkroadBot.UI            # WPF Desktop Application
├── plugins/
│   ├── SilkroadBot.Plugin.Chat           # Chat monitoring & auto-reply
│   ├── SilkroadBot.Plugin.CommandCenter  # Command-based bot control
│   ├── SilkroadBot.Plugin.Logger         # Event logging to file
│   └── SilkroadBot.Plugin.Protection     # Auto-heal, flee, death recovery
├── docs/
│   ├── BUILD.md
│   └── PLUGIN_DEVELOPMENT.md
└── Build/                        # Compiled output directory

✨ Features

Core Framework

• Packet-driven architecture - Full protocol handling for Silkroad Online
• Dual execution modes - Clientless or attach-to-running-client
• Multi-region support - iSRO, TRSRO, vSRO, and custom private servers
• Async/await throughout - Non-blocking operations for responsiveness
• Dependency Injection - Clean, testable, extensible design

Plugin System

• Dynamic DLL loading - Drop plugins into the plugins/ folder
• Rich event hooks - Connection, packets, chat, inventory, movement, party, skills
• Complete isolation - Plugins can't modify core logic
• Plugin SDK - Full SDK with base classes, attributes, and interfaces

AI Engine (Gemini)

• IAIProvider interface - Pluggable AI backends
• GeminiProvider - Google Generative AI integration
• Command Validation Layer - Every AI decision passes through safety guardrails
• Per-profile AI config - Different settings for each character
• Confidence thresholds - Reject uncertain AI decisions automatically

Safety Guardrails

• Death prevention
• Health threshold enforcement
• Position safety validation
• Combat overload prevention
• Confidence scoring
• Action whitelist validation

Navigation

• A* pathfinding - Grid-based navigation with diagonal support
• Path smoothing - Line-of-sight optimization
• Map data provider - Pluggable map data loading
• NavMesh-ready - Abstract interface for future algorithms

UI

• Modern WPF - Dark-themed Windows desktop application
• Dashboard - Live status overview
• Profile Manager - Create, edit, duplicate profiles
• Packet Monitor - Real-time packet inspection
• Plugin Manager - Enable/disable plugins
• AI Settings - Configure AI per profile
• Live Logs - Filtered real-time logging

🚀 Quick Start

Prerequisites

• Windows 10/11
• Visual Studio 2022 (with .NET desktop workload)
• .NET 8.0 SDK

Build

# Using the build script
.\build.ps1

# Or via dotnet CLI
dotnet build SilkroadBot.sln -c Release -o Build/

Run

# Run the UI application
dotnet run --project src/SilkroadBot.UI/

Develop Plugins

See PLUGIN_DEVELOPMENT.md for the complete plugin development guide.

🔌 Supported Regions

🤖 AI Integration

The AI layer is optional and requires a Gemini API key. Configure via the UI or profile JSON:

{
  "ai": {
    "enabled": true,
    "providerName": "Gemini",
    "apiKey": "YOUR_API_KEY",
    "model": "gemini-pro",
    "temperature": 0.3,
    "decisionIntervalMs": 2000
  }
}

Every AI decision passes through the Command Validation Layer:

1. Action existence check
2. Confidence threshold
3. Death prevention
4. Health safety
5. Combat safety
6. Position validity

📋 Project Status

• Core architecture & networking
• Protocol adapters (iSRO, TRSRO, vSRO)
• Plugin system with SDK
• Event dispatch system
• AI integration with guardrails
• Pathfinding (A*)
• Profile management
• WPF UI
• Sample plugins (4)
• Documentation

⚠️ Disclaimer

This project is intended for educational purposes and private server use. It does not include anti-cheat evasion techniques. Use responsibly and in accordance with the terms of service of any game server you connect to.

📄 License

MIT License - See LICENSE for details.

<!-- ml-intern-provenance -->

Generated by ML Intern

This model repository was generated by ML Intern, an agent for machine learning research and development on the Hugging Face Hub.

• Try ML Intern: https://smolagents-ml-intern.hf.space
• Source code: https://github.com/huggingface/ml-intern

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = 'Ahmedramadan24/SilkroadBot'
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

For non-causal architectures, replace AutoModelForCausalLM with the appropriate AutoModel class.