Image Processing MCP Server

Instructions

Implementation Reference

• tools/filters.py:439-487 (handler)

  Core handler function that validates input, loads image using ImageProcessor, applies PIL's ImageFilter.FIND_EDGES, generates output image data, and returns JSON response.

  async def apply_find_edges(image_data: str) -> list[TextContent]:
      """
      应用边缘检测滤镜
      
      Args:
          image_data: 图片数据（base64编码）
          
      Returns:
          应用滤镜后的图片数据
      """
      try:
          # 验证参数
          if not image_data:
              raise ValidationError("图片数据不能为空")
          
          # 加载图片
          image = processor.load_image(image_data)
          
          # 应用边缘检测滤镜
          edges_image = image.filter(ImageFilter.FIND_EDGES)
          
          # 输出处理后的图片
          output_info = processor.output_image(edges_image, "find_edges")
          
          result = {
              "success": True,
              "message": "边缘检测滤镜应用成功",
              "data": {
                  **output_info,
                  "filter_type": "find_edges",
                  "size": image.size
              }
          }
          
          return [TextContent(type="text", text=json.dumps(result, ensure_ascii=False))]
          
      except ValidationError as e:
          error_result = {
              "success": False,
              "error": f"参数验证失败: {str(e)}"
          }
          return [TextContent(type="text", text=json.dumps(error_result, ensure_ascii=False))]
          
      except Exception as e:
          error_result = {
              "success": False,
              "error": f"边缘检测滤镜应用失败: {str(e)}"
          }
          return [TextContent(type="text", text=json.dumps(error_result, ensure_ascii=False))]

• main.py:312-325 (registration)

  MCP tool registration using FastMCP's @mcp.tool() decorator. Provides input schema via Pydantic Annotated Field and delegates execution to the core handler in filters.py.

  @mcp.tool()
  def apply_find_edges(
      image_source: Annotated[str, Field(description="图片源，可以是文件路径或base64编码的图片数据")]
  ) -> str:
      """应用边缘检测滤镜"""
      try:
          result = safe_run_async(filters_apply_find_edges(image_source))
          return result[0].text
      except Exception as e:
          return json.dumps({
              "success": False,
              "error": f"应用边缘检测失败: {str(e)}"
          }, ensure_ascii=False, indent=2)

• tools/filters.py:107-120 (schema)

  Explicit JSON schema definition for the tool input in get_filter_tools() function, though not directly used in main MCP registration.

  Tool(
      name="apply_find_edges",
      description="应用边缘检测滤镜",
      inputSchema={
          "type": "object",
          "properties": {
              "image_data": {
                  "type": "string",
                  "description": "图片数据（base64编码）"
              }
          },
          "required": ["image_data"]
      }
  ),