import * as $dara from '@darabonba/typescript'; export declare class SubmitAIAgentVideoAuditTaskRequestCallbackConfig extends $dara.Model { /** * @remarks * The authentication token for callback. * * @example * Bearer Token */ token?: string; /** * @remarks * The URL for receiving callback notifications. * * @example * https://yourcallback */ url?: string; static names(): { [key: string]: string; }; static types(): { [key: string]: any; }; validate(): void; constructor(map?: { [key: string]: any; }); } export declare class SubmitAIAgentVideoAuditTaskRequestCapturePolicies extends $dara.Model { /** * @remarks * The duration over which to capture the specified number of frames. Unit: seconds. * * @example * 10 */ duration?: number; /** * @remarks * The number of frames to capture. * * @example * 2 */ frameCount?: number; /** * @remarks * The text prompt to send to the MLLM along with the captured frames. */ prompt?: string; /** * @remarks * The timestamp in the video at which to start capturing frames. Unit: seconds. * * @example * 0 */ startTime?: number; static names(): { [key: string]: string; }; static types(): { [key: string]: any; }; validate(): void; constructor(map?: { [key: string]: any; }); } export declare class SubmitAIAgentVideoAuditTaskRequestInput extends $dara.Model { /** * @remarks * The OSS URL of the input file. Format: * * http(s)://{BucketName}.{Endpoint}/{ObjectName} * * @example * http://my-bucket.cn-shanghai.aliyuncs.com/object-id.mp4 */ media?: string; /** * @remarks * The type of the input file. Valid values: * * * OSS: an OSS object. * * @example * OSS */ type?: string; static names(): { [key: string]: string; }; static types(): { [key: string]: any; }; validate(): void; constructor(map?: { [key: string]: any; }); } export declare class SubmitAIAgentVideoAuditTaskRequest extends $dara.Model { /** * @remarks * The ID of the AI agent. * * This parameter is required. * * @example * *******3b3d94abda22****** */ AIAgentId?: string; /** * @remarks * The interval, in milliseconds, at which to submit captured frames to the AI agent. Valid values: 0 to 5000. Default value: 3000. If it is set to 0, all captured frames are sent to the model in a single batch request. Otherwise, frames are sent sequentially with the specified interval between each request. * * @example * 3000 */ auditInterval?: number; /** * @remarks * Callback configurations. * * @example * {"Url":"https://yourcallback","Token":"yourtoken"} */ callbackConfig?: SubmitAIAgentVideoAuditTaskRequestCallbackConfig; /** * @remarks * An array of frame-capturing policies. Each policy defines a set of frames to be analyzed and will generate a separate result from the model. * * This parameter is required. */ capturePolicies?: SubmitAIAgentVideoAuditTaskRequestCapturePolicies[]; /** * @remarks * The details of the input file. * * This parameter is required. */ input?: SubmitAIAgentVideoAuditTaskRequestInput; /** * @remarks * The user-defined data. */ userData?: string; static names(): { [key: string]: string; }; static types(): { [key: string]: any; }; validate(): void; constructor(map?: { [key: string]: any; }); }