VoidMuse

---

🌐 Live Demo

• Online Web: https://voidmuse-dev.github.io/voidmuse/

---

🎯 Project Vision

More than just an AI plugin, it's an AI engineering learning platform

VoidMuse is committed to building efficient and intelligent AI IDE plugins through open source component integration with minimal development costs. We believe in the power of open source, enabling every developer to easily build their own AI tools through carefully selected and integrated excellent open source components.

🌟 Core Philosophy

• 📚 Learning-Oriented: Provides comprehensive technical documentation to help users deeply understand AI engineering
• 🔧 Open Source Integration: Based on 20+ excellent open source components, lowering development barriers
• 🧠 Knowledge Transfer: Comprehensive coverage of cutting-edge AI concepts from Embedding to vector databases
• 💡 Practice-Driven: Combining theory with practice, learning while doing

---

🚀 Quick Start

Usage Methods

Direct plugin installation

VS Code Installation

1. Open VS Code Extension Marketplace (Ctrl+Shift+X)
2. Search for "VoidMuse"
3. Click install and restart the editor

IntelliJ IDEA Installation

1. Open Settings → Plugins
2. Search for "VoidMuse"
3. Click install and restart IDE

For Learners - Local development startup details see Development Guide

🎯 3-Minute Quick Setup

Want to experience VoidMuse's powerful features immediately? Follow our quick configuration guide and get started in three steps!

👉 📖 View 3-Step Quick Setup Guide

💡 Just configure the AI model API key to start using basic features. Search and code understanding features can be configured later as needed.

Detailed Configuration Guide

If you need more detailed configuration instructions, check out the following documentation:

• 🤖 AI Model Configuration - Get free API keys and configuration guide
• 🔍 Search Function Configuration - Google Search configuration detailed tutorial
• 🔍 Bocha Search Configuration - Domestic search service configuration
• 🧠 Code Understanding Configuration - Use the same AI model API

---

🏗️ Overall Architecture

VoidMuse adopts a modular architecture design, supports multi-IDE platforms, integrates multiple AI models, and provides a complete AI development experience.

Detailed architecture documentation: Technical Architecture Description

---

🔬 Technical Highlights & Learning Value

🧠 AI Engineering Core Concepts

• Embedding Technology: Complete analysis from principles to practice
• Vector Database: Local indexing and retrieval mechanisms
• Context Management: Context optimization strategies for large model conversations
• MCP Protocol: Standardized implementation of multimodal conversation protocols

🔧 Open Source Component Ecosystem

We carefully select and integrate the following excellent open source components:

• Ant Design 5.x - Enterprise-class UI design language
• TipTap - Rich text editor
• Vercel's ai-sdk - Multi-model support, agent framework

📖 Learning Resources Summary

1. Microsoft AI Agents for Beginners: Basic knowledge for building AI agents, https://learn.microsoft.com/en-us/shows/ai-agents-for-beginners/
2. Andrew Ng's Agentic AI Course: Practical applications of agentic AI, https://deeplearning.ai
3. Hugging Face AI Agents Course: Practical AI agent development skills, https://huggingface.co/learn/agents-course/en/unit0/introduction
4. Google 5-Day AI Agent Training: Hands-on project experience, https://rsvp.withgoogle.com/events/google-ai-agents-intensive_2025/home
5. Anthropic Official Course: Advanced AI Agent technology, https://anthropic.skilljar.com
6. Coursera AI Agents Specialization: Combining academic theory with practical applications, https://coursera.org/specializations/ai-agents
7. Salesforce AI Agent Course: Enterprise-level AI Agent solutions, https://salesforce.com/ap/agentforce/ai-agent-course/

---

🛠️ Features

🧠 Intelligent Code Completion (AutoComplete)

• Technical Principles: Context-aware intelligent completion algorithms
• Learning Value: Understanding the implementation mechanisms of AI code completion
• Core Features:
  • Real-time code analysis
  • Context-aware completion
  • Multi-language support

🔍 Local Codebase Indexing (Codebase)

• Technical Principles: Vectorized storage + semantic retrieval
• Learning Value: Master the construction and optimization of vector databases
• Core Features:
  • Local vector database
  • Semantic similarity retrieval
  • Intelligent code context extraction

💬 Multi-Model AI Chat (AI Chat)

• Technical Principles: Multi-turn dialogue + tool calling + context management
• Learning Value: Understanding engineering practices of large model applications
• Supported Models:
  • OpenAI GPT series
  • Anthropic Claude series
  • DeepSeek series
  • More models supported through OpenRouter

🔗 MCP Protocol Support

• Technical Principles: Standardized multimodal conversation protocol
• Learning Value: Design patterns for cross-platform AI interaction
• Protocol Features:
  • Standardized tool calling
  • Multimodal data processing
  • Extensible plugin architecture

🔍 Intelligent Search Integration

• Supported Engines: Google Search, Bocha AI Search
• Configuration Documentation:
  • Google Search Configuration
  • Bocha Search Configuration
• Free Resources: Large Model Provider Free Quota Summary

---

📚 Learning Path Guide

🎯 Target Audience

• AI engineers wanting to learn IDE plugin development
• Frontend/backend developers wanting to learn AI integration
• Students and researchers wanting to understand AI engineering
• Open source enthusiasts wanting to participate in AI projects

📖 Recommended Learning Path

Beginner Path (AI Concept Introduction)

1. What is Embedding?
2. Vector Database Basics

🛠️ Practice Projects

• Implementing Deep Search
• Codebase Feature Technical Deep Analysis
• Custom Embedding Model Integration
• Extending MCP Protocol Support
• Implementing Custom Search Engines

---

🚀 Future Feature Roadmap

Feature Module	Core Features	Technical Principles	Learning Value
🧠 Context Memory	Cross-session memory, intelligent prioritization, personalized learning	Vector database + semantic retrieval	AI memory system design
🗜️ Context Compression	Dynamic length management, multi-level compression, semantic preservation	Intelligent summarization + hierarchical compression	Large model context optimization
🤖 Agent Code Generation	Requirement analysis, architecture design, automatic implementation, test validation	Multi-Agent collaboration framework	AI-driven software development

---

📊 Data Collection Notice

🔒 Privacy Protection Commitment

To better understand user usage and improve product experience, VoidMuse integrates minimalist tracking functionality in the GUI interface. We promise:

• ✅ Only collect page access data: Only record page entry and exit times to calculate page dwell time
• ✅ No sensitive information collection: No collection of any code content, personal information, or operation details
• ✅ Transparent open source: All tracking code is completely open source, viewable in gui/src/services/AnalyticsService.ts

📚 Tracking Technology Learning Value

Our tracking implementation is also a great frontend data analysis learning case:

• 🎯 Multi-platform support: Simultaneously integrates Google Analytics and Baidu Statistics
• ⏱️ Precise timing: Accurate calculation methods for page dwell time
• 🔄 Lifecycle management: Complete handling of page visibility changes
• 🛡️ Exception handling: Graceful degradation when tracking fails

---

🤝 Contributing Guide

We welcome all forms of contributions! Whether it's bug fixes, feature development, documentation improvements, or technical sharing, all can make VoidMuse better.

Detailed contribution guide: Contributing Guide

Community Contributors

Thanks to all developers who contribute code, documentation, and ideas to the project. Your contributions make VoidMuse better!

📞 Contact Us

• 🐛 Issue Reporting: GitHub Issues
• 📧 Email Contact: voidmuse@qq.com

💬 Community Communication

---

📄 License

This project is licensed under the Apache License 2.0 - see the LICENSE file for details.