Agent Scripting API 400 on Bulk Import via Terraform Module

Stumbled on a weird bug today with the Agent Scripting API when attempting to bulk import scripts using our custom Terraform module in the AU-1 BYOC environment. The Terraform apply command completes without errors, but the scripts do not appear in the console, and the debug logs show a 400 Bad Request response from the POST /api/v2/interaction/configs/scripts endpoint. The payload includes valid JSON with version=2, type=script, and a properly formatted sections array containing text and clickToDial blocks. The error message returned is "message":"Validation failed: Section ID 'intro_001' is already in use or malformed.", yet the ID is unique within the current deployment scope. We are using the Genesys Cloud Terraform provider v1.68.4 and the GC CLI for validation. The script works perfectly when created manually via the UI, suggesting a schema mismatch or a hidden dependency in the API contract that is not documented. The environment variables are set correctly for GENESYS_CLOUD_REGION=au-1 and authentication uses a service account with interaction:scripts:write permissions. Has anyone encountered similar validation errors when deploying scripts via code? The HCL configuration looks standard: resource "genesyscloud_script" "onboarding_script" { name = "Onboarding Flow v2" description = "Automated onboarding script" version = 2 ... }. The issue persists across multiple runs, even after deleting the script manually and re-running the apply. The timestamp in the logs indicates the request is reaching the backend, but the validation layer is rejecting it silently without detailed field-level errors. We need to automate this for our CI/CD pipeline, so manual fixes are not an option. Any insights on the expected JSON structure for sections or known issues with bulk imports in the current provider version? The timezone settings in the script metadata are set to Australia/Sydney, which should be valid. We are also seeing occasional 429 Too Many Requests errors when retrying, which complicates debugging. The network latency to AU-1 is normal, around 5ms. We have checked the API documentation for interaction/configs/scripts but found no specific examples for bulk operations with complex section types. The error seems to point to a conflict, but the ID is definitely unique in the current tenant context. We are considering switching to direct API calls via GitHub Actions if Terraform support remains unstable, but prefer to keep everything in HCL for consistency. Any workarounds or known fixes for this validation error?