Screenshot Server
by KunihiroS
This project provides an MCP server designed to capture screenshots and facilitate their use by other processes or AI assistants. It primarily saves the screenshot to a file path specified by the client (Host).
Last updated: N/A
What is Screenshot Server?
The Screenshot Server is an MCP (Modular Communication Protocol) server that captures screenshots and saves them to a specified file path, enabling other processes or AI assistants to easily access and utilize the captured images.
How to use Screenshot Server?
The server provides several tools to capture and save screenshots. The recommended workflow involves the AI assistant (Host) providing its workspace path, allowing the server to save the screenshot directly into the Host's file system. Alternatively, screenshots can be saved to a fixed directory or an arbitrary path.
Key features of Screenshot Server
Saves screenshots to specified file paths
Supports WSL to Windows workflows using UNC paths
Provides multiple tools for different use cases
Integrates with MCP for communication with AI assistants
Configurable save locations and filenames
Use cases of Screenshot Server
Capturing screenshots for AI assistant analysis
Automating screenshot capture in workflows
Saving screenshots to a shared workspace for collaboration
Integrating screenshot capture into automated testing processes
Providing a screenshot service for other applications
FAQ from Screenshot Server
Where does the server save the screenshots?
Where does the server save the screenshots?
The server can save screenshots to the Host's workspace, a fixed 'images/' directory, or an arbitrary path specified by the client.
How do I configure the server to work with WSL?
How do I configure the server to work with WSL?
The server must run on Windows. Configure your MCP Host (running in WSL) to launch the server on Windows using PowerShell and provide the correct Windows paths.
What are the prerequisites for running the server?
What are the prerequisites for running the server?
You need Python 3.x and the required dependencies (mcp[cli]>=1.4.1, pyautogui, Pillow) installed on the Windows machine where the server will run.
What is the recommended workflow for using this server with an AI assistant?
What is the recommended workflow for using this server with an AI assistant?
The recommended workflow is to use the save_screenshot_to_host_workspace tool, providing the AI assistant's WSL workspace path as an argument.
How do I install the dependencies?
How do I install the dependencies?
Use uv sync to install the required libraries defined in pyproject.toml.