file-organizer-MCP-Server
If you are the rightful owner of file-organizer-MCP-Server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.
File Organizer MCP Server uses Google's Gemini Vision API to analyze screenshots and organize files based on extracted order.
File Organizer MCP Server
A Model Context Protocol (MCP) server that uses Google's Gemini Vision API to analyze screenshots of file listings/course contents and automatically organize files based on the extracted order.
Features
- Screenshot Analysis: Uses Gemini Vision API to extract file organization order from screenshots
- Smart File Organization: Matches files with extracted order and renames them with numeric prefixes
- Descriptive Folders: Creates section-specific folders based on content type
- AI/OCR Handling: Robust handling of common OCR variations (e.g., AI vs Al)
- Undo Support: Can revert organized files back to their original state
Prerequisites
- Python 3.13+
- Visual Studio Code
- Google API key for Gemini Vision API
- MCP Extension for VS Code
uv
package manager (recommended) or pip
Setup
- Clone the repository:
git clone https://github.com/mutaician/file-organizer-MCP-Server.git
cd file-organizer-mcp
- Create and activate virtual environment using uv:
uv venv
source .venv/bin/activate # On Windows: .venv\Scripts\activate
uv pip install -r requirements.txt
-
Install VS Code MCP Extension:
- Open VS Code
- Go to Extensions (Ctrl+Shift+X)
- Install the GitHub Copilot extension (or any extension that supports MCP servers).
-
Configure VS Code workspace:
- Create
.vscode
folder if it doesn't exist - Add
mcp.json
with the following content:
{ "inputs": [ { "type": "promptString", "id": "gemini-key", "description": "Gemini API Key", "password": true } ], "servers": { "File Organizer": { "type": "stdio", "command": "${workspaceFolder}/.venv/bin/python", "args": ["${workspaceFolder}/src/server.py"], "env": { "GOOGLE_API_KEY": "${input:gemini-key}" } } } }
- Create
Usage in VS Code
-
Start the MCP Server:
- Open Command Palette (Ctrl+Shift+P)
- Type "MCP: List Servers"
- Select "File Organizer"
- Enter your Gemini API key when prompted
- You should now see your server and its tools available in the Copilot Chat/Agent mode.
-
In the Copilot Chat, you can organize your files by:
- First, analyze a screenshot to get the file order:
Could you analyze the screenshot at "/path/to/screenshot.jpg"?
- Then organize files using the extracted order:
Could you organize the files in "/path/to/section" using this order?
- The server will automatically:
- Create a descriptive subfolder based on the section title
- Number files according to their order
- Handle AI/OCR variations
- First, analyze a screenshot to get the file order:
-
Additional Commands:
- List files in a directory:
Could you list the files in "/path/to/directory"?
- Undo organization:
Could you undo the organization from "/path/to/organized" back to "/path/to/source"?
- List files in a directory:
-
Best Practices:
- Keep screenshots clear and well-lit for better OCR
- Verify organization results after each operation
- Use the undo function if needed before trying again
Key Concepts
-
Screenshot Analysis:
- Takes a screenshot of your desired file organization (e.g., course content listing)
- Uses Gemini Vision to extract the order and hierarchy
- Handles numbered lists and various formats
-
File Organization:
- Matches files to the extracted order
- Adds numeric prefixes (e.g., "01_", "02_")
- Creates descriptive subfolders based on content type
- Preserves original filenames after the prefix
-
Smart Matching:
- Handles variations in OCR text (AI vs Al)
- Ignores case sensitivity
- Removes special characters that might interfere
- Matches partial names when appropriate
Troubleshooting
-
If files aren't matching correctly:
- Check for special characters in filenames
- Verify the screenshot text is clear and readable
- Ensure file extensions are correct
-
If the server won't start:
- Verify your Gemini API key is valid
- Check Python virtual environment is activated
- Ensure all dependencies are installed
Dependencies
mcp-sdk
: Model Context Protocol implementationgoogle-generativeai
: Gemini AI API clientpillow
: Image processing for screenshots