JianYing MCP

Instructions

Implementation Reference

• jianyingdraft/tool/video_tool.py:294-358 (handler)

  Primary MCP tool handler for 'add_video_keyframe'. Decorated with @mcp.tool() for registration. Includes input schema via type hints and detailed docstring. Retrieves draft/track info and delegates to service layer.
  @mcp.tool() def add_video_keyframe( video_segment_id: str, property_name: str, time_offset: str, value: float ) -> ToolResponse: """ 为视频片段添加关键帧 Args: video_segment_id: 视频片段ID，通过add_video_segment获得 property_name: 属性名称，可选参数如下： position_x：右移为正, 此处的数值应该为`剪映中显示的值` / `草稿宽度`, 也即单位是半个画布宽 position_y：上移为正, 此处的数值应该为`剪映中显示的值` / `草稿高度`, 也即单位是半个画布高 rotation：顺时针旋转的**角度** scale_x：单独控制X轴缩放比例(1.0为不缩放), 与`uniform_scale`互斥 scale_y：单独控制Y轴缩放比例(1.0为不缩放), 与`uniform_scale`互斥 uniform_scale：同时控制X轴及Y轴缩放比例(1.0为不缩放), 与`scale_x`和`scale_y`互斥 alpha：不透明度, 1.0为完全不透明, 仅对`VideoSegment`有效 saturation：饱和度, 0.0为原始饱和度, 范围为-1.0到1.0 contrast：对比度, 0.0为原始对比度, 范围为-1.0到1.0 brightness：亮度, 0.0为原始亮度, 范围为-1.0到1.0 volume：音量, 1.0为原始音量 time_offset: 时间偏移量，格式如 "0.5s", "1s" 等 value: 属性值 Examples: # 在0.5秒时设置水平位置 add_video_keyframe("video_segment_id", "position_x", "0.5s", 0.2) # 在1秒时设置旋转角度 add_video_keyframe("video_segment_id", "rotation", "1s", 45.0) # 在2秒时设置透明度 add_video_keyframe("video_segment_id", "alpha", "2s", 0.5) """ # 通过video_segment_id获取相关信息 draft_id = index_manager.get_draft_id_by_video_segment_id(video_segment_id) track_info = index_manager.get_track_info_by_video_segment_id(video_segment_id) if not draft_id: return ToolResponse( success=False, message=f"未找到视频片段ID对应的草稿: {video_segment_id}" ) if not track_info: return ToolResponse( success=False, message=f"未找到视频片段ID对应的轨道信息: {video_segment_id}" ) track_name = track_info.get("track_name") # 调用服务层处理业务逻辑 result = add_video_keyframe_service( draft_id=draft_id, video_segment_id=video_segment_id, property_name=property_name, time_offset=time_offset, value=value, track_name=track_name ) return result
• jianyingdraft/services/video_service.py:254-326 (helper)

  Supporting service function that performs the core logic: instantiates VideoSegment and invokes its add_keyframe method, handles exceptions, and formats ToolResponse.
  def add_video_keyframe_service( draft_id: str, video_segment_id: str, property_name: str, time_offset: str, value: float, track_name: Optional[str] = None ) -> ToolResponse: """ 视频关键帧添加服务 - 为视频片段添加关键帧 Args: draft_id: 草稿ID video_segment_id: 视频片段ID property_name: 属性名称，如 "position_x", "rotation", "alpha" 等 time_offset: 时间偏移量，如 "0.5s", "1s" 等 value: 属性值 track_name: 轨道名称（可选） Returns: ToolResponse: 包含操作结果的响应对象 """ try: # 创建VideoSegment实例，传入video_segment_id video_segment = VideoSegment(draft_id, video_segment_id=video_segment_id, track_name=track_name) # 调用视频关键帧添加方法 result_data = video_segment.add_keyframe( property_name=property_name, time_offset=time_offset, value=value ) # 构建返回数据 response_data = { "video_segment_id": video_segment_id, "draft_id": draft_id, "property_name": property_name, "time_offset": time_offset, "value": value, "add_keyframe": result_data } # 如果有轨道名称，添加到返回数据中 if track_name: response_data["track_name"] = track_name return ToolResponse( success=True, message=f"视频关键帧添加成功: {property_name}={value} at {time_offset}", data=response_data ) except ValueError as e: # 处理参数错误 return ToolResponse( success=False, message=f"参数错误: {str(e)}" ) except NameError as e: # 处理轨道不存在错误 return ToolResponse( success=False, message=f"轨道错误: {str(e)}" ) except Exception as e: # 处理其他未预期的错误 return ToolResponse( success=False, message=f"视频关键帧添加失败: {str(e)}" )