Loading source
Pulling the file list, source metadata, and syntax-aware rendering for this listing.
Source from repo
Build and deploy AI applications on Azure AI Foundry using Microsoft's model catalog and AI services
Files
Skill
Size
Entrypoint
Format
Open file
Syntax-highlighted preview of this file as included in the skill package.
resource/private-network/references/custom-template-adaptation.md
1# Custom Template Adaptation23For the EXTEND path — when the user has existing Bicep or Terraform templates.45## Instructions671. **Read** the user's existing template files. Understand the resource graph: what's defined, how resources reference each other, what naming conventions are used.892. **Analyze** the template against the user's requirements (from [intake.md](intake.md)) and the Foundry private networking documentation validated in the intake step. Identify:10- Resources already present and correctly configured11- Resources present but misconfigured (wrong settings, missing properties)12- Resources missing entirely13- Dependency or wiring issues (e.g., PEs referencing wrong subnet, DNS zones not linked)14153. **Present** findings to the user as a gap analysis table: resource, status (✅ present / ⚠️ misconfigured / ❌ missing), and what needs to change. Include any issues found.16174. **Propose** an end-to-end plan to address all gaps — ordered by dependency. Explain what will be added, what will be modified, and why. Never overwrite existing modules — add alongside and reference existing resources.18195. **Wait** for user approval before making any changes.20216. **Implement** the approved changes. After implementation, the flow continues to Step 4 (Pre-Deployment Validation) in the main workflow.2223## Retry Safety2425> ⚠️ If a deployment fails after the capability host step starts, Azure Container Apps leaves a `legionservicelink` service association on the agent subnet that **cannot be removed**. On retry, use a **new subnet or new VNet** — never reuse the same agent subnet.2627