Looking for some advice on troubleshooting this validation failure we are encountering while deploying a new bot flow via the AppFoundry integration suite. The issue manifests specifically when the bot attempts to transition from a self-service task to a live agent transfer. We are using the standard /api/v2/analytics/prompts endpoint to validate the prompt configurations before pushing the updated bot definition to production.
The payload includes a standard transfer task with a defined queue and a fallback prompt. However, the API consistently returns a 400 Bad Request response, citing an issue with the prompt object structure. This is perplexing because the same JSON structure works perfectly in a standalone org we use for testing. The discrepancy suggests there might be a hidden dependency on org-level settings or a specific version of the Prompt API schema that is not documented in the standard reference.
Here is the exact error response we are receiving:
{
"code": "bad.request",
"message": "Prompt configuration invalid: missing required field 'text' or 'audioUrl' in transfer node",
"messageParams": [],
"errors": [
{
"code": "invalid.prompt",
"message": "Transfer node requires a valid prompt definition"
}
]
}
We have verified that the text field is present and non-empty in our payload. We are also checking against the latest API documentation for version 2.0. The environment is a multi-org setup where the partner integration manages configuration across several client tenants. Is there a known issue with how the Prompts API handles transfer nodes in specific org configurations? We suspect this might be related to how the API validates prompt dependencies when the bot is linked to a specific IVR context. Any insights into the internal validation logic for transfer prompts would be greatly appreciated.