mcp-openvision

#!/usr/bin/env python """ Basic example of using the OpenVision MCP server. This example shows how to: 1. Connect to the server 2. List available tools 3. Call the image_analysis tool with a simple prompt """ import asyncio import os from pathlib import Path from mcp import ClientSession, StdioServerParameters from mcp.client.stdio import stdio_client async def main(): # Set up the server parameters server_params = StdioServerParameters( command="python", args=["-m", "mcp_openvision"], env={ # You can set environment variables here "OPENROUTER_API_KEY": os.environ.get("OPENROUTER_API_KEY"), "OPENROUTER_DEFAULT_MODEL": "qwen/qwen2.5-vl-32b-instruct:free", # Optional: set default model }, ) print("Connecting to OpenVision MCP server...") # Connect to the server async with stdio_client(server_params) as (read_stream, write_stream): async with ClientSession(read_stream, write_stream) as session: # Initialize the connection await session.initialize() # List available tools tools_response = await session.list_tools() print(f"Available tools: {[tool.name for tool in tools_response.tools]}") # Read an image file image_path = Path(__file__).parent / "sample_image.jpg" if not image_path.exists(): print(f"Error: Sample image not found at {image_path}") print( "Please add a sample image named 'sample_image.jpg' to the examples directory." ) return with open(image_path, "rb") as f: image_data = f.read() print("\nAnalyzing image with basic prompt...") # Call the image_analysis tool with a simple query result = await session.call_tool( "image_analysis", {"image": {"data": image_data}, "query": "What's in this image?"}, ) print(f"\nResult:\n{result.content}") # Example with a detailed contextual query print("\nAnalyzing image with a detailed contextual query...") result = await session.call_tool( "image_analysis", { "image": {"data": image_data}, "query": "Analyze this photograph for use in a travel blog. Identify notable landmarks, architectural styles, and cultural elements that would interest tourists. Suggest what time of day it was taken and how lighting affects the mood of the image. This will help readers understand what to expect when visiting this location." }, ) print(f"\nResult with detailed contextual query:\n{result.content}") # Example with custom system prompt print("\nAnalyzing image with custom system prompt...") result = await session.call_tool( "image_analysis", { "image": {"data": image_data}, "query": "Identify all colors present in this image and explain how they work together in the composition", "system_prompt": "You are a color analysis expert with deep knowledge of color theory and visual design. Focus on identifying precise color values, their relationships, emotional impacts, and practical applications in design contexts. Analyze colors using terminology from professional color systems (RGB, CMYK, HSL, etc.) when relevant." }, ) print(f"\nResult with custom system prompt:\n{result.content}") # Example with relative path and project_root print("\nAnalyzing image using a relative path with project_root...") # Get the current working directory to use as a project_root example project_root = os.getcwd() # Call the image_analysis tool with a relative path and project_root try: result = await session.call_tool( "image_analysis", { "image": "examples/test_image.png", # Relative path from project root "project_root": project_root, # Set the project root directory "query": "What can you see in this image?", }, ) print( f"\nResult with relative path and project_root:\n{result.content}" ) except Exception as e: print(f"\nError using relative path: {str(e)}") print( "Make sure you have a file at 'examples/test_image.png' relative to the current directory." ) print("\nDone!") if __name__ == "__main__": # Run the async main function asyncio.run(main())