tos_image_process

Process images stored in TOS buckets by applying operations like resize, format conversion, crop, rotate, and quality adjustment with specified parameters.

Instructions

图片处理（组合操作，支持多种处理参数）

Input Schema

TableJSON Schema

Name	Required	Description
`bucket_name`	Yes	存储桶名称
`object_key`	Yes	图片对象键名
`process`	Yes	图片处理参数。参数格式通常为 'image/操作,参数'，如: 'image/resize,h_100' 或 'image/format,jpg'。常用操作包括：resize(缩放),format(格式转换),quality(质量),crop(裁剪),rotate(旋转)等。
`save_bucket`	Yes	保存的存储桶名称
`save_key`	Yes	保存的对象键名

Implementation Reference

src/tos_mcp_server/handlers.py:218-260 (handler)

Core handler function that executes the image processing using TOS SDK's get_object method with process, save_bucket, and save_object parameters for persistent processing.

async def image_process(args: Dict[str, Any]) -> List[TextContent]:
    """图片处理（支持持久化）"""
    bucket_name = args["bucket_name"]
    object_key = args["object_key"]
    process = args["process"]
    save_bucket = args["save_bucket"]
    save_key = args["save_key"]
    
    try:
        # 使用官方SDK写法，通过save_bucket和save_object参数执行图片处理和持久化
        resp = tos_client.get_object(
            bucket=bucket_name,
            key=object_key,
            process=process,
            save_bucket=base64.b64encode(save_bucket.encode("utf-8")).decode("utf-8"),
            save_object=base64.b64encode(save_key.encode("utf-8")).decode("utf-8")
        )
        
        # 读取处理结果以确保处理完成
        processed_data = resp.read()
        
        # 等待一下确保回写完成
        import time
        time.sleep(1.0)
        
        # 生成处理后对象的预签名 URL
        download_url = tos_client.pre_signed_url(tos.HttpMethodType.Http_Method_Get, save_bucket, save_key, 3600)
        
        result = {
            "presigned_url": download_url.signed_url,
            "source_bucket": bucket_name,
            "source_key": object_key,
            "save_bucket": save_bucket,
            "save_key": save_key,
            "process": process,
            "processed_size": len(processed_data),
            "expires_in": 3600,
            "status": "processed"
        }
        return [TextContent(type="text", text=json.dumps(result, indent=2, ensure_ascii=False))]
    except Exception as e:
        return [TextContent(type="text", text=f"图片处理失败: {str(e)}")]

src/tos_mcp_server/server.py:244-273 (registration)

Tool registration in list_tools() including name, description, and input schema definition.

Tool(
    name="tos_image_process",
    description="图片处理（组合操作，支持多种处理参数）",
    inputSchema={
        "type": "object",
        "properties": {
            "bucket_name": {
                "type": "string",
                "description": "存储桶名称"
            },
            "object_key": {
                "type": "string",
                "description": "图片对象键名"
            },
            "process": {
                "type": "string",
                "description": "图片处理参数。参数格式通常为 'image/操作,参数'，如: 'image/resize,h_100' 或 'image/format,jpg'。常用操作包括：resize(缩放),format(格式转换),quality(质量),crop(裁剪),rotate(旋转)等。"
            },
            "save_bucket": {
                "type": "string",
                "description": "保存的存储桶名称"
            },
            "save_key": {
                "type": "string",
                "description": "保存的对象键名"
            }
        },
        "required": ["bucket_name", "object_key", "process", "save_bucket", "save_key"]
    }
),

src/tos_mcp_server/server.py:355-356 (registration)
Dispatch logic in call_tool() that routes tos_image_process calls to the image_process handler.
```
elif name == "tos_image_process":
    return await image_process(arguments)
```

src/tos_mcp_server/server.py:16-20 (registration)

Import of the image_process handler function from handlers module.

    create_bucket, list_buckets, get_bucket_meta, delete_bucket,
    put_object, get_object, list_objects, delete_object,
    presigned_url, image_process, image_info,
    video_snapshot, video_info
)

TOS MCP Server