blend_images
Combine two images using blend modes like normal, multiply, screen, or overlay with adjustable opacity to create composite visuals.
Instructions
混合两张图片
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| image1_source | Yes | 第一张图片源，可以是文件路径或base64编码的图片数据 | |
| image2_source | Yes | 第二张图片源，可以是文件路径或base64编码的图片数据 | |
| blend_mode | No | 混合模式：normal（正常）、multiply（正片叠底）、screen（滤色）、overlay（叠加） | normal |
| opacity | No | 第二张图片的不透明度，范围 0.0-1.0 | |
| output_format | No | 输出格式：PNG、JPEG、WEBP 等 | PNG |
Output Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| result | Yes |
Implementation Reference
- tools/advanced.py:722-836 (handler)Core handler implementing image blending: loads two images, resizes them according to mode, applies opacity to second image, uses blend modes (normal, multiply, etc.), and outputs base64 result.
async def blend_images(arguments: Dict[str, Any]) -> List[TextContent]: """ 混合两张图片 Args: arguments: 包含两张图片源和混合参数的字典 Returns: List[TextContent]: 处理结果 """ try: # 参数验证 image1_source = arguments.get("image1_source") image2_source = arguments.get("image2_source") ensure_valid_image_source(image1_source) ensure_valid_image_source(image2_source) blend_mode = arguments.get("blend_mode", "normal") opacity = arguments.get("opacity", 0.5) resize_mode = arguments.get("resize_mode", "fit_first") output_format = arguments.get("output_format", DEFAULT_IMAGE_FORMAT) # 验证参数 validate_numeric_range(opacity, 0.0, 1.0, "opacity") processor = ImageProcessor() # 加载图片 image1 = processor.load_image(image1_source) image2 = processor.load_image(image2_source) # 转换为RGBA模式 if image1.mode != "RGBA": image1 = image1.convert("RGBA") if image2.mode != "RGBA": image2 = image2.convert("RGBA") # 调整尺寸 if resize_mode == "fit_first": image2 = image2.resize(image1.size, Image.Resampling.LANCZOS) final_size = image1.size elif resize_mode == "fit_second": image1 = image1.resize(image2.size, Image.Resampling.LANCZOS) final_size = image2.size elif resize_mode == "fit_largest": if image1.width * image1.height > image2.width * image2.height: image2 = image2.resize(image1.size, Image.Resampling.LANCZOS) final_size = image1.size else: image1 = image1.resize(image2.size, Image.Resampling.LANCZOS) final_size = image2.size else: # fit_smallest if image1.width * image1.height < image2.width * image2.height: image2 = image2.resize(image1.size, Image.Resampling.LANCZOS) final_size = image1.size else: image1 = image1.resize(image2.size, Image.Resampling.LANCZOS) final_size = image2.size # 调整第二张图片的透明度 alpha_channel = image2.split()[-1] alpha_channel = alpha_channel.point(lambda p: int(p * opacity)) image2.putalpha(alpha_channel) # 应用混合模式 if blend_mode == "normal": result = Image.alpha_composite(image1, image2) elif blend_mode == "multiply": # 简化的乘法混合 result = Image.blend(image1, image2, opacity) elif blend_mode == "screen": # 简化的屏幕混合 result = Image.blend(image1, image2, opacity) else: # 其他混合模式使用普通混合 result = Image.alpha_composite(image1, image2) # 转换为base64 output_info = processor.output_image(result, "batch_resize", output_format) return [TextContent( type="text", text=json.dumps({ "success": True, "message": f"成功混合图片，使用{blend_mode}模式", "data": { **output_info, "metadata": { "size": f"{result.width}x{result.height}", "blend_mode": blend_mode, "opacity": opacity, "resize_mode": resize_mode, "format": output_format } } }, ensure_ascii=False) )] except ValidationError as e: return [TextContent( type="text", text=json.dumps({ "success": False, "error": f"参数验证失败: {str(e)}" }, ensure_ascii=False) )] except Exception as e: return [TextContent( type="text", text=json.dumps({ "success": False, "error": f"混合图片失败: {str(e)}" }, ensure_ascii=False) )] - main.py:726-749 (registration)Tool registration via @mcp.tool() decorator in main.py. Defines input parameters with descriptions and defaults matching the handler schema, constructs arguments dict, and calls the advanced handler via safe_run_async.
@mcp.tool() def blend_images( image1_source: Annotated[str, Field(description="第一张图片源，可以是文件路径或base64编码的图片数据")], image2_source: Annotated[str, Field(description="第二张图片源，可以是文件路径或base64编码的图片数据")], blend_mode: Annotated[str, Field(description="混合模式：normal（正常）、multiply（正片叠底）、screen（滤色）、overlay（叠加）", default="normal")], opacity: Annotated[float, Field(description="第二张图片的不透明度，范围 0.0-1.0", ge=0.0, le=1.0, default=0.5)], output_format: Annotated[str, Field(description="输出格式：PNG、JPEG、WEBP 等", default="PNG")] ) -> str: """混合两张图片""" try: arguments = { "image1_source": image1_source, "image2_source": image2_source, "blend_mode": blend_mode, "opacity": opacity, "output_format": output_format } result = safe_run_async(advanced_blend_images(arguments)) return result[0].text except Exception as e: return json.dumps({ "success": False, "error": f"混合图片失败: {str(e)}" }, ensure_ascii=False, indent=2) - tools/advanced.py:197-239 (schema)JSON schema definition for blend_images tool inputs within the Tool object in get_advanced_tools(), detailing properties, enums, defaults, and required fields.
Tool( name="blend_images", description="混合两张图片", inputSchema={ "type": "object", "properties": { "image1_source": { "type": "string", "description": "第一张图片源（文件路径或base64编码）" }, "image2_source": { "type": "string", "description": "第二张图片源（文件路径或base64编码）" }, "blend_mode": { "type": "string", "description": "混合模式", "enum": ["normal", "multiply", "screen", "overlay", "soft_light", "hard_light"], "default": "normal" }, "opacity": { "type": "number", "description": "第二张图片的透明度（0.0-1.0）", "minimum": 0.0, "maximum": 1.0, "default": 0.5 }, "resize_mode": { "type": "string", "description": "尺寸调整模式", "enum": ["fit_first", "fit_second", "fit_largest", "fit_smallest"], "default": "fit_first" }, "output_format": { "type": "string", "description": "输出格式", "enum": ["PNG", "JPEG", "WEBP"], "default": "PNG" } }, "required": ["image1_source", "image2_source"] } ),