Signal Cloud Notifications and Webhooks
The API contract covers Discord notification endpoints, generic webhook endpoints, delivery history, and replay of failed deliveries. These operations require an authenticated Console session and declare webhook or delivery permissions individually.
First read-only check
With a deployment-authorized session that has webhooks.read, call GET /v1/notifications/discord/endpoints using the deployment host. A 200 response contains an endpoints array; use the operation's limit parameter only when you need to change the documented page size. This checks access without creating or changing an endpoint.
Endpoint lifecycle
The contract includes operations to list and create Discord or generic webhook endpoints, inspect them, create revisions, enable or disable them, archive them, and rotate a generic webhook secret. A newly created Discord endpoint starts disabled. Creating a generic webhook returns its generated secret once; rotating a secret also returns the new value once. The API description says destination credentials are not returned by endpoint detail operations.
Treat the one-time secret response as sensitive. The published contract does not include an endpoint for retrieving a secret later. Use the request and response schemas in the API reference before implementing endpoint management.
If you create a generic webhook, capture its returned secret at that moment and store it using your deployment's approved secret handling. Do not put a real secret into a guide, source file, or shared log. A newly created Discord endpoint starts disabled; follow the documented enable operation only after its destination has been checked.
Delivery history
The API publishes delivery-list and delivery-detail operations, plus replay for failed deliveries. Delivery reads require deliveries.read; replay requires deliveries.replay and is limited to failed deliveries by the documented operation.
Troubleshooting
- For a
403, confirm whether the operation needswebhooks.read,webhooks.write,deliveries.read, ordeliveries.replay; ask the deployment's access administrator if the session lacks the required community permission. - For a
404, check that the endpoint or delivery identifier belongs to the active community selected by the session. - For a
409, review the operation's current-state and idempotency details before retrying a write. - If a delivery is not successful, inspect its documented delivery detail and state. Replay is only defined for failed deliveries; this API does not document how to repair the external destination itself.