What is the reason the Predictive Routing configuration fails when attempting to map Zendesk ticket tags to Genesys Cloud skills? We are migrating a support queue that relies heavily on tag-based routing in Zendesk, and we are trying to replicate this logic using Genesys Cloud skills for predictive distribution. The goal is to ensure that interactions tagged with ‘billing’ in Zendesk route to agents with the ‘Billing_L2’ skill in Genesys.
When testing the migration script using the Genesys Cloud Java SDK (version 13.1.0), the POST request to /v2/routing/skills returns a 400 Bad Request. The error indicates that the skill name contains characters that are invalid for Genesys Cloud, even though these tags are perfectly valid in Zendesk. Specifically, we are using hyphens in our Zendesk tags, such as ‘billing-issue’, which seem to cause the conflict.
Here is the error response we are receiving:
{
"message": "Invalid skill name. Skill names must contain only alphanumeric characters, underscores, and spaces.",
"code": "bad_request"
}
We have tried replacing hyphens with underscores in our mapping logic, but the predictive routing model still fails to assign the correct skill priority. The Architect flow seems to accept the interaction, but the routing queue does not prioritize it based on the mapped skill. This is causing a significant bottleneck in our migration timeline, as we need to preserve the existing routing logic from Zendesk.
Is there a specific character encoding issue we need to address? Or is there a recommended workaround for mapping Zendesk tags that contain special characters to Genesys Cloud skills? We are based in Europe/Paris, and our timezone settings are correctly configured in both systems, so we suspect this is purely a data mapping issue rather than a regional configuration problem. Any advice on best practices for cleaning tag data before migration would be greatly appreciated.