Trigger and retrieve Agora Cloud Recording to Amazon S3

Verified steps

1. Enable Cloud Recording in Agora Console for your project; obtain a Customer ID and Customer Secret from the RESTful API section to authenticate Cloud Recording REST calls using HTTP Basic Auth.
2. Call the acquire endpoint (POST to the Agora Cloud Recording API acquire path) with your appId, channelName, uid (a unique string not used by any participant), and clientRequest containing scene and resourceExpiredHour; the response returns a resourceId valid for a short window.
3. Immediately call the start endpoint with the resourceId, mode (e.g., 'mix' for composite or 'individual' for per-track), and a clientRequest body that includes token, storageConfig (with vendor, region, bucket, and access credential fields populated from your S3 setup), and recordingConfig.
4. Poll the query endpoint with resourceId and sid to monitor recording status; the response includes fileList showing files uploaded to S3 so far and the current status.
5. Call the stop endpoint when the session ends, passing resourceId, sid, and uid; Agora finalizes the recording, uploads remaining segments to S3, and returns the final fileList.
6. Configure an S3 bucket policy and IAM credentials granting Agora's recording service s3:PutObject access; Agora documents the specific IAM permissions required in their Cloud Recording storage configuration guide.