Chris4K commited on
Commit
e169048
Β·
verified Β·
1 Parent(s): cd95b99

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +72 -2
README.md CHANGED
@@ -1,12 +1,82 @@
1
  ---
2
- title: Mcp Images
3
- emoji: 🐠
4
  colorFrom: gray
5
  colorTo: red
6
  sdk: gradio
7
  sdk_version: 5.34.0
8
  app_file: app.py
9
  pinned: false
 
 
 
 
10
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
11
 
12
  Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
 
1
  ---
2
+ title: Mcp Images Server
3
+ emoji: πŸ–ΌοΈ
4
  colorFrom: gray
5
  colorTo: red
6
  sdk: gradio
7
  sdk_version: 5.34.0
8
  app_file: app.py
9
  pinned: false
10
+ tags:
11
+ - images
12
+ - mcp-server
13
+
14
  ---
15
+
16
+
17
+ # Image Analysis MCP Server
18
+
19
+ This Gradio application serves as an MCP (Model Control Protocol) server that provides image analysis tools for LLMs like Claude.
20
+
21
+ ## πŸ”§ Available Tools
22
+
23
+ - **analyze_image**: Comprehensive image analysis including dimensions, format, colors, and orientation
24
+ - **get_image_orientation**: Determine if an image is portrait, landscape, or square
25
+ - **count_colors**: Analyze color information and identify dominant colors
26
+ - **extract_text_info**: Basic analysis for potential text content in images
27
+
28
+ ## πŸš€ Usage with Claude Desktop
29
+
30
+ To use this MCP server with Claude Desktop, add this configuration to your `claude_desktop_config.json`:
31
+
32
+ ```json
33
+ {
34
+ "mcpServers": {
35
+ "image-analysis": {
36
+ "url": "https://YOUR_USERNAME-image-analysis-mcp.hf.space/gradio_api/mcp/sse"
37
+ }
38
+ }
39
+ }
40
+ ```
41
+
42
+ Replace `YOUR_USERNAME` with your actual HuggingFace username.
43
+
44
+ ## πŸ“ Example Prompts
45
+
46
+ Once configured, you can ask Claude:
47
+
48
+ - "Analyze this image for me"
49
+ - "What are the dominant colors in this photo?"
50
+ - "Is this image portrait or landscape?"
51
+ - "Does this image contain text?"
52
+
53
+ Claude will automatically use these tools to provide detailed image analysis!
54
+
55
+ ## πŸ” How It Works
56
+
57
+ 1. Upload an image to Claude or paste an image
58
+ 2. Claude automatically converts the image to base64
59
+ 3. Claude calls the appropriate MCP tool on this server
60
+ 4. The server processes the image using PIL (Python Imaging Library)
61
+ 5. Results are returned to Claude and presented to you
62
+
63
+ ## πŸ› οΈ Technical Details
64
+
65
+ - Built with **Gradio** and **PIL (Pillow)**
66
+ - Accepts images in various formats (JPEG, PNG, GIF, etc.)
67
+ - Automatically handles base64 image data from MCP clients
68
+ - Provides JSON-formatted analysis results
69
+ - Includes error handling for invalid inputs
70
+
71
+ ## πŸ“‹ Requirements
72
+
73
+ - `gradio[mcp]>=4.0.0`
74
+ - `Pillow>=9.0.0`
75
+
76
+ ## πŸ”— Links
77
+
78
+ - [MCP Documentation](https://modelcontextprotocol.io/)
79
+ - [Gradio MCP Guide](https://www.gradio.app/guides/building-an-mcp-server-with-gradio)
80
+ - [Claude Desktop](https://claude.ai/desktop)
81
 
82
  Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference