🚀 FileGen MCP Server

FileGen MCP Server is a lightweight Model Context Protocol (MCP) server built to automate file handling, project initialization, and command execution — enabling seamless integration with LLM-based coding workflows.

🧠 Why FileGen?

When LLMs generate code and project structures, the developer often has to manually create folders, files, and paste code — a repetitive and time-consuming process.

FileGen MCP Server eliminates this friction by letting agents or tools directly:

• Create files/folders
• Write and delete content
• Generate project structures
• Execute terminal commands

It’s the perfect backend utility for building smarter, faster, AI-assisted development pipelines.

---

✨ Features

• 📂 Read Files: Read individual files or recursively scan directories.
• 🏗️ Initialize Projects: Generate complete directory structures from templates.
• 📝 Write Files: Create or append content to files.
• ❌ Delete Files: Remove unnecessary files or folders.
• 💻 Execute Commands: Run system commands via Python subprocess.

---

🛠️ Available Tools

🔹 read_files

Reads one or more files/directories from the local filesystem.

Parameters:

• paths (List[str]): List of file/directory paths to read.

---

🔹 init_project

Generates a folder/file structure.

Parameters:

• name (str): Project name.
• path (str, optional): Base path (default: system root directory).
• structure (Dict[str, Any]): Nested dictionary defining folders/files.

---

🔹 write_file

Creates or appends content to a file.

Parameters:

• path (str): Target file path.
• content (str): Content to write.
• mode (str, optional): "w" (overwrite) or "a" (append). Default: "w".

---

🔹 delete_path

Deletes a file or directory.

Parameters:

• path (str): Target path to delete.

---

🔹 execute_command

Executes a shell command.

Parameters:

• params (Dict[str, Any]): Command and arguments to execute.

---

⚙️ Installation

1. Install dependencies:

   uv sync
   

2. Start the server:

   uv run python src/main.py
   

---

📜 Bash Script Setup

Use the setup_and_run.sh script to configure and launch the MCP server in one go.

✅ Steps:

1. Ensure Python 3 is installed.

2. Make the script executable:

   chmod +x setup_and_run.sh
   

3. Run the script:

   ./setup_and_run.sh
   

🧰 What It Does:

• Verifies Python and uv availability.
• Installs uv if missing.
• Sets up a virtual environment using uv.
• Installs dependencies from pyproject.toml.
• Generates a default config.json.
• Launches the server.

---

🧩 Configuration

A config.json file is auto-generated when you run the setup script.

To manually integrate with your agent system, ensure the cwd field points to the correct project directory.

💡 Tip: Include the project path in your LLM prompt so it knows where to write files.

Example:

{
  "mcpServers": {
    "project": {
      "command": "uv",
      "args": ["run", "python", "src/main.py"],
      "cwd": "/your/absolute/path/project",
      "env": {
        "PYTHONPATH": "/your/absolute/path/project/src"
      }
    }
  }
}

---

📁 Project Structure

project/
├── src/
│   ├── main.py                 # MCP server entry point
│   ├── actions/
│   │   ├── read_files.py       # Read files/directories
│   │   ├── init_project.py     # Project structure creation
│   │   ├── delete_file.py      # File/folder deletion
│   │   └── execute_command.py  # System command execution
│   └── schemas/
│       └── project_structure.py # Pydantic models for project 
└── setup_and_run.sh            # Bootstrap script