Webhook Notifications


Several operations of the Evidence Manager API run asynchronously: the HTTP request only starts the process, and the API notifies you through a webhook once it finishes. You can configure one or more callback URLs to receive these notifications.

Every notification is sent as an HTTP POST request with a JSON body with the following shape:

{
  "operation": "EVIDENCE_GROUP_CLOSED",
  "result": "OK",
  "data": {
    "evidenceGroupId": "4c33124e-d444-4e97-bbaa-50375c364aca",
    "caseFileId": "3fa85f64-5717-4562-b3fc-2c963f66afa6"
  }
}
  • operation: identifies the event that triggered the notification. Possible values: EVIDENCE_CREATION, EVIDENCE_DELETE, EVIDENCE_GROUP_CLOSED, REPORT_CREATION and SIGNATURE_CREATION.
  • result: currently always OK.
  • data: contains the details of the event. Its content depends on the value of operation, as described below.

Evidence created

operation: "EVIDENCE_CREATION". Sent when an "Evidence" finishes its registration process (hash timestamping and, for internally custodied files, the upload to storage).

{
  "operation": "EVIDENCE_CREATION",
  "result": "OK",
  "data": {
    "evidenceId": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
    "evidenceGroupId": "4c33124e-d444-4e97-bbaa-50375c364aca",
    "caseFileId": "57081c08-08e7-4d7e-b10a-8a98d7b1b3bd"
  }
}

evidenceGroupId and caseFileId are only present when the "Evidence" was registered within an "Evidence Group" / "Case file".

Evidence deleted

operation: "EVIDENCE_DELETE". Sent when the file of an "Evidence" has been fully deleted from storage.

{
  "operation": "EVIDENCE_DELETE",
  "result": "OK",
  "data": {
    "evidenceId": "3fa85f64-5717-4562-b3fc-2c963f66afa6"
  }
}

Evidence Group closed

operation: "EVIDENCE_GROUP_CLOSED". Sent when an "Evidence Group" reaches the CLOSED status after a close request.

{
  "operation": "EVIDENCE_GROUP_CLOSED",
  "result": "OK",
  "data": {
    "evidenceGroupId": "4c33124e-d444-4e97-bbaa-50375c364aca",
    "caseFileId": "3fa85f64-5717-4562-b3fc-2c963f66afa6"
  }
}

Report generated

operation: "REPORT_CREATION". Sent when a signed report has been generated and timestamped.

{
  "operation": "REPORT_CREATION",
  "result": "OK",
  "data": {
    "reportId": "1d6d735d-0383-42d1-bfb9-6d4cf71aa739",
    "reportName": "example-report",
    "reportUrl": "https://example.com/reports/example-report.pdf",
    "evidencePackageUrl": "https://example.com/reports/example-report.zip"
  }
}

Sign file completed

operation: "SIGNATURE_CREATION". Sent when the Embed company seal to pdf process finishes.

{
  "operation": "SIGNATURE_CREATION",
  "result": "OK",
  "data": {
    "processId": "1d6d735d-0383-42d1-bfb9-6d4cf71aa739",
    "fileId": "3fa85f64-5717-4562-b3fc-2c963f66afa6",
    "status": "FILE_SIGNED_FINISHED",
    "url": "https://example.com/signed/example-report.pdf"
  }
}

Configuration & Security

Webhook notifications are configured per tenant, by us, as part of your deployment. There is no endpoint to create, list or change them: ask your EADFactory contact.

You can provide one or more callback destinations; when more than one is configured, every one of them receives every notification.

What you give us is a base, not a full URL: scheme, host and port. Any path you include in it is dropped.

Your endpoint has to expose these fixed paths, one per kind of notification:

POST /evidences/callback
POST /evidence-groups/callback
POST /certificates/callback
POST /signatures/callback

So a base of https://example.com receives evidence notifications at https://example.com/evidences/callback.

Every webhook request includes an Authorization header with the token agreed with your EADFactory contact:

Authorization: Bearer <your_token>

This lets your endpoint verify the authenticity of the request by checking the token.

Delivery and retries

A notification is considered delivered when your endpoint responds with 200, 201, 204, 404 or 422.

A 400 or a 401 is not retried. Both are treated as permanent errors and the notification is dropped there and then, with no further attempt and no queue to recover it from. If your endpoint rejects a notification because your token expired, or because a body did not validate, that notification is lost.

Only server errors and connection failures are retried, with increasing delays and a limited number of attempts.

Because of this, your endpoint should be idempotent, should not assume notifications arrive strictly once or in order, and should answer 200 as soon as it has accepted the payload, doing the work afterwards.