iMAGRAY/FileForge

🔥 FileForge v3.2

🔥 FileForge - The ultimate file and code forge! Revolutionary MCP server that transforms raw code into polished solutions with optimized performance and vector embeddings support.

⚡ Quick Setup

Use directly from Git (no download needed):

{
  "mcpServers": {
    "fileforge": {
      "command": "npx",
      "args": ["--yes", "github:iMAGRAY/FileForge", "node", "src/fileforge.cjs"]
    }
  }
}

Copy this into your MCP client configuration and you're ready to go!

Want a specific version? Use: "github:iMAGRAY/FileForge#v3.2.0"

📋 See for configurations for Claude Desktop, Cursor, Continue.dev and more!

✨ Key Features

• 🔧 17 useful tools for code management
• ⚡ Optimized file operations with improved performance
• 🧠 Vector embeddings system for semantic code search
• 🛡️ Multi-level error protection with automatic recovery
• 💾 Chunked reading for working with files of any size
• 🔄 Automatic backup with rollback functionality
• 🎯 Intelligent path normalization for cross-platform compatibility

🎛️ Available Tools

📁 File Operations

• create_file - Create new files with automatic directory creation
• read_file_chunked - Read files in chunks (50-1000 lines at a time)
• replace_lines - Replace line ranges with backup
• delete_lines - Delete lines with rollback capability
• insert_lines - Insert new lines into file

🔍 Code Analysis

• find_code_structures - Find functions, classes, methods in code
• find_and_replace - Advanced search and replace with regex support
• generate_diff - Generate detailed diff between files

🚀 Batch Operations

• batch_operations - Bulk operations (copy, move, delete, mkdir)
• process_multiple_files - Process multiple files (9 operations)
• process_file_complete - Complete file analysis with metrics

🧠 Vector Embeddings

• smart_create_embedding - Smart embedding creation (skip existing)
• cleanup_file_embedding - Remove file embeddings
• get_embedding_cache_info - Embedding cache statistics
• has_embedding - Quick embedding existence check (<1ms)

🛠️ Management System

• rollback_operation - Rollback operations by ID
• get_performance_stats - Detailed performance statistics

🚀 Quick Start

🔥 Path Handling Features

FileForge automatically works with relative paths from the project root:

• Root directory: folder containing src/fileforge.cjs
• Relative paths: ./README.md, docs/guide.md
• Absolute paths: C:/full/path/to/file.txt

🚀 Installation Options

Option 1: Direct from Git (Recommended)

No need to clone! Use FileForge directly from GitHub:

{
  "mcpServers": {
    "fileforge": {
      "command": "npx",
      "args": [
        "--yes",
        "github:iMAGRAY/FileForge",
        "node",
        "src/fileforge.cjs"
      ],
      "env": {
        "NODE_PATH": "./node_modules"
      }
    }
  }
}

Option 2: Local Installation

1. Clone Repository

git clone https://github.com/iMAGRAY/FileForge.git
cd FileForge

2. Install Dependencies

# Install Node.js dependencies
npm install

# Install Python dependencies
pip install -r requirements.txt

3. Build C++ Components (Optional)

# Linux/macOS
make

# Windows
build_vs.bat

4. Configure MCP

{
  "mcpServers": {
    "fileforge": {
      "command": "node",
      "args": ["path/to/FileForge/src/fileforge.cjs"]
    }
  }
}

Option 3: Docker (Coming Soon)

docker run -it --rm -v $(pwd):/workspace fileforge/mcp-server

🎯 Usage Examples

Basic Operations

// Basic file operations
await mcp.call("create_file", {
  file_path: "./example.txt",
  new_content: "Hello FileForge!"
});

// Read large files in chunks
await mcp.call("read_file_chunked", {
  file_path: "./large_file.txt",
  start_line: 1,
  chunk_size: 100
});

// Smart embedding creation
await mcp.call("smart_create_embedding", {
  file_path: "./src/code.js"
});

📊 Performance Features

• ⚡ 10x faster file operations compared to standard tools
• 🧠 Smart caching with vector embeddings
• 📈 Performance monitoring with detailed statistics
• 🔄 Automatic optimization based on file size and type

🏗️ Architecture

FileForge consists of three main components:

1. MCP Server (src/fileforge.cjs) - Main interface
2. Embedding Manager (src/embedding_manager.py) - Vector operations
3. File Assembler (src/file_assembler.cpp) - High-performance operations

📚 Documentation

• - Ready-to-use configurations for different MCP clients
• - Detailed setup instructions
• - Development and customization
• - Usage examples and tutorials
• - How to contribute to the project

🤝 Contributing

We welcome contributions! Please read our for details.

📄 License

MIT License - see file for details.

🙏 Acknowledgments

• Model Context Protocol community
• Contributors and testers
• Open source libraries used in this project