Image Processing MCP Server

Instructions

Implementation Reference

• tools/filters.py:489-537 (handler)

  Core implementation of the apply_smooth tool handler. Loads the image from base64 data, applies PIL's ImageFilter.SMOOTH filter, generates output info using ImageProcessor, and returns JSON result.

  async def apply_smooth(image_data: str) -> list[TextContent]:
      """
      应用平滑滤镜
      
      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)
          
          # 应用平滑滤镜
          smooth_image = image.filter(ImageFilter.SMOOTH)
          
          # 输出处理后的图片
          output_info = processor.output_image(smooth_image, "smooth")
          
          result = {
              "success": True,
              "message": "平滑滤镜应用成功",
              "data": {
                  **output_info,
                  "filter_type": "smooth",
                  "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"平滑滤镜应用失败: {str(e)}"
          }
          return [TextContent(type="text", text=json.dumps(error_result, ensure_ascii=False))]

• main.py:327-338 (registration)

  MCP tool registration for apply_smooth using @mcp.tool() decorator. Wraps the filters.apply_smooth handler with safe_run_async and error handling, defining input schema via Annotated Field.

  def apply_smooth(
      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:
      """应用平滑滤镜"""
      try:
          result = safe_run_async(filters_apply_smooth(image_source))
          return result[0].text
      except Exception as e:
          return json.dumps({
              "success": False,
              "error": f"应用平滑效果失败: {str(e)}"
          }, ensure_ascii=False, indent=2)

• tools/filters.py:121-134 (schema)

  Tool schema definition for apply_smooth in get_filter_tools(), specifying inputSchema requiring 'image_data' string.

  Tool(
      name="apply_smooth",
      description="应用平滑滤镜",
      inputSchema={
          "type": "object",
          "properties": {
              "image_data": {
                  "type": "string",
                  "description": "图片数据（base64编码）"
              }
          },
          "required": ["image_data"]
      }
  ),

• main.py:40-51 (registration)

  Import of the apply_smooth handler from tools.filters as filters_apply_smooth for use in main.py tool registration.

  from tools.filters import (
      apply_blur as filters_apply_blur,
      apply_gaussian_blur as filters_apply_gaussian_blur,
      apply_sharpen as filters_apply_sharpen,
      apply_edge_enhance as filters_apply_edge_enhance,
      apply_emboss as filters_apply_emboss,
      apply_find_edges as filters_apply_find_edges,
      apply_smooth as filters_apply_smooth,
      apply_contour as filters_apply_contour,
      apply_sepia as filters_apply_sepia,
      apply_invert as filters_apply_invert
  )