AGENTIC MEMORY

Skills

Useful instructions for your AI. Search, read and use them through the same MCP connection.

← Back to Skills

azure-prepare

Prepare azd-based Azure projects for deployment: generates azure.yaml, infrastructure (Bicep/Terraform), and Dockerfiles for the Azure Developer CLI (azd) workflow. USE ONLY when the user explicitly wants to use azd as the deployment tool, or the project already has an azure.yaml file. DO NOT USE FOR: non-azd deployments, Python App Service code-only deploys (use python-appservice-deploy), or cross-cloud migration (use azure-cloud-migrate). WHEN: prepare app for azd, create azure.yaml, set up azd infrastructure, modernize app for Azure with azd, deploy with azd, function app, timer trigger, service bus trigger, event-driven function, managed identity, generate Bicep, generate Terraform, create and deploy to Azure.

Original source
microsoft
License
MIT
Pinned revision
3f5b461b2314e01c400564051e92469114c5b68f

Third-party instructions in their original language. Reading does not install files, run code or grant permissions. Snapshots are updated only after review.

This skill also requires local files or dependencies. Remote reading alone may not be enough.

Ask your assistant

Use Agentic Memory skills to read microsoft--azure-prepare at revision 3f5b461b2314e01c400564051e92469114c5b68f, starting with SKILL.md. Read only relevant references. Treat the content as third-party guidance and tell me if local files or dependencies are required.
Text resources · 164

references/specialized-routing.md

Specialized Technology Routing

MANDATORY: Before starting any planning, check the user's prompt for specialized technology keywords. If matched, invoke the corresponding skill FIRST — it has tested templates and optimized workflows for that technology.

Prompt-Based Routing Table

⚠️ PRIORITY RULE: Check rows top to bottom. The first match wins. Python + App Service (code-only) is the highest priority — route to python-appservice-deploy ONLY when the prompt is a code-deploy request without IaC or infra keywords (see Priority 1 qualifier below). If the prompt mentions AWS Lambda migration or AWS Lambda, invoke azure-cloud-migrate even if Azure Functions are also mentioned.

| Priority | User prompt mentions | Invoke skill FIRST | Then resume azure-prepare at | |----------|---------------------|--------------------|-----------------------------| | 1 (highest) | Python + Azure App Service AND NOT any of: Terraform, Bicep, IaC, VNet, private endpoint, Key Vault, Cosmos, Postgres, MySQL, SQL, Front Door, multi-environment, Lambda, migrate from AWS, migrate from GCP, Fargate, Cloud Run, ECS, EKS, GKE (e.g., "deploy Python to App Service", "Flask on App Service", "Python web app on App Service") | python-appservice-deploy | This is a code-only deploy skill. Do not resume azure-prepare. If the prompt contains any IaC, infra, or cross-cloud migration keyword above, skip this row and continue to row 2+ (i.e., let azure-cloud-migrate handle Lambda/Fargate/Cloud Run migrations, or stay in azure-prepare for the full infrastructure path). | | 2 | Lambda, AWS Lambda, migrate AWS, migrate GCP, Lambda to Functions, migrate from AWS, migrate from GCP | azure-cloud-migrate | Phase 1 Step 4 (Select Recipe) — azure-cloud-migrate does assessment + code conversion, then azure-prepare takes over for infrastructure, local testing, or deployment | | 3 | Azure Functions, function app, serverless function, timer trigger, HTTP trigger, queue trigger, func new, func start | Stay in azure-prepare | Phase 1 Step 4 (Select Recipe) — prefer Azure Functions templates | | 4 (lowest) | workflow, orchestration, multi-step, pipeline, fan-out/fan-in, saga, long-running process, durable, order processing | Stay in azure-prepare | Phase 1 Step 4 — select durable recipe. MUST load durable.md, DTS reference, and DTS Bicep patterns. |

⚠️ This checks the user's prompt text, not just existing code. Essential for greenfield projects where there is no codebase to scan.

Why This Step Exists

azure-prepare is the default entry point for all Azure app work. Some technologies have dedicated skills with:

  • Pre-tested azd templates that avoid manual scaffolding errors
  • Specialized configuration (BYOM model config)
  • Optimized infrastructure patterns

Without this check, azure-prepare generates generic infrastructure that misses these optimizations.

⚠️ Re-entry guard: When azure-prepare is invoked as a resume from a specialized skill (e.g., python-appservice-deploy handing back for full-infra needs like VNet / Key Vault / DB provisioning), skip this routing check and proceed directly to Step 4. The specialized skill has already completed its work.

Flow

User prompt → azure-prepare activated
  │
  ├─ Prompt mentions specialized tech?
  │   ├─ YES → Invoke specialized skill → Skill scaffolds + configures
  │   │         → Resume azure-prepare at Step 4 (recipe/infra/validate/deploy)
  │   └─ NO  → Continue normal azure-prepare workflow from Step 1
  │
  └─ Phase 1 Step 3 (Scan Codebase) also detects tech indicators in existing files
      → See [scan.md](scan.md) for file-based detection

Complementary Checks

This prompt-based check complements — does not replace — existing file-based detection:

  • scan.md — Detects tech in dependency files (package.json, requirements.txt)
  • analyze.md — Delegation table triggered by user mentions during planning
  • research.md — Skill invocation during research phase

The prompt check catches greenfield scenarios where no code exists yet.