| name | azure-gcp-de-para-article |
| description | Add or review an Azure-to-GCP service equivalence ("de-para") section in a technical article on orafaelferreira.com — a comparison table plus practical differences, framed as trade-offs rather than marketing claims. Use when writing or updating an article whose primary subject is an Azure resource/service/architecture, or when reviewing Azure-centric technical content for a multi-cloud comparison. |
Azure↔GCP De-Para for Articles
Ported from ai/skills/azure-gcp-de-para-article.SKILL.md into Claude Code skill format.
Pairs with article-authoring-site-context (editorial rules) and mcp-research-orchestrator
(sourcing) — use all three together when writing an Azure-centric article.
When to use
- Writing an article about an Azure resource, service, or architecture
- Updating an existing Azure article to add a multi-cloud comparison
- Reviewing Azure-centric technical text to make GCP equivalences explicit
When not to use
- Articles unrelated to cloud
- Purely opinion pieces with no technical substance
- Comparisons that would require a formal benchmark you don't have real data for
Goal
Every article centered on an Azure service should include a clear de-para with GCP,
highlighting real equivalences, behavioral differences, and the limits of the comparison —
never implying full parity where there isn't any.
Before writing
Load article-authoring-site-context's editorial rules, ai/mcp/source-of-truth-policy.md
for sourcing rules, and TECH_STACK.md for project context.
Required rules
- For every Azure resource that's a primary subject of the article, map the GCP equivalent.
- Always flag partial equivalence — never claim full parity.
- Explain differences across at least 3 dimensions: operation, pricing, technical limits.
- Include official links inline at the first mention of both the Azure and GCP service.
- Prefer primary sources: official Azure docs and official Google Cloud docs.
- Avoid marketing language; focus on technical decision-making and trade-offs.
Recommended article structure
- Problem context on Azure
- Azure solution (with examples)
## De-para Azure and GCP section- Practical implementation differences
- Risks, lock-in, and operational costs
- Conclusion with a decision criterion
De-para section template
## De-para Azure and GCP
| Scenario | Azure | GCP | Practical notes |
|---|---|---|---|
| Serverless compute | Azure Functions | Cloud Functions | Trigger model, timeout, concurrency, and cold start differ |
| Managed containers | Azure Container Apps | Cloud Run | Scale-to-zero, networking, and observability vary |
| Managed Kubernetes | Azure Kubernetes Service (AKS) | Google Kubernetes Engine (GKE) | Autopilot/operation mode and base cost differ |
| Messaging | Azure Service Bus | Pub/Sub | Delivery semantics, ordering, and dead-lettering differ |
| Managed relational DB | Azure SQL Database | Cloud SQL | HA, replicas, and tuning differ per engine |
| Observability | Azure Monitor + Application Insights | Cloud Monitoring + Cloud Logging | Collection model and integration differ |
| Identity and access | Microsoft Entra ID + RBAC | Cloud IAM | Permission scope and governance model differ |
Quality checklist
- Explicit
## De-para Azure and GCP section is present
- Every central Azure service in the text has a GCP counterpart or a stated justification for its absence
- Each de-para row states whether the equivalence is full, partial, or approximate only
- Official links are present for every service cited
- The text describes at least one migration risk or lock-in consideration
Completion hints
| field | guidance |
|---|
category | prefer Artigos when it's an in-depth technical comparison |
tags | include Azure, GCP, Cloud, Arquitetura, plus the specific topic |
excerpt | explicitly mention that the article includes an Azure×GCP de-para |
Technical rigor notes
- Never assume equivalence from product naming alone.
- If GCP has multiple options for one Azure resource, list the primary one and mention alternatives.
- When there's no direct equivalent, state
no direct equivalent and suggest a service composition instead.
