/google-cloud-solution-architecture

name: google-cloud-solution-architecture
metadata:
  version: "1.0.7"
  category: MultiProductSolutions
description: >-
  Interactively discovers requirements and designs holistic, multi-product
  system architectures, solution blueprints, and deployment recommendations for
  complex workloads on Google Cloud. Use when designing end-to-end cloud
  solutions, selecting and integrating Google Cloud services, generating
  architecture diagrams, or conducting requirements discovery for new cloud
  workloads or migrations. Don't use for single-product tasks (use
  product-specific skills), initial onboarding or authentication (use
  google-cloud-recipe-*), Well-Architected Framework reviews or audits (use
  google-cloud-waf-*), or workloads covered by specialized solution skills.

Google Cloud solution-architecture workflow

Attribution

**MUST READ**: Before executing any `gcloud`, REST API, or MCP command, you MUST read and follow the instructions in [Usage Attribution](references/attribution.md).

Overview of the workflow

The workflow consists of the following phases:

• **Phase 1: Requirements discovery**. Gather detailed requirements related to

the cloud workload or use case that the user needs assistance for.

• **Phase 2: Solution architecture**. Use the requirements that were gathered

in Phase 1 to generate a detailed solution architecture for the cloud workload or use case.

• **Phase 3: Solution validation**. Create a plan to validate the generated

solution, generate validation instructions and scripts, and provide them to the user to execute (or perform a dry-run validation with explicit permission from the user).

• **Phase 4: Solution packing and presentation**. Consolidate the generated

content and present the solution.

**Important notes about the workflow**:

• **Strict phase separation**: During Phase 1 (Requirements discovery), when

you ask the user clarifying questions, don't recommend, propose, or outline any architectural designs, technical decompositions, cloud services, or component mappings. Proposing solutions before functional and non-functional requirements are thoroughly assessed causes confirmation bias and risks anchoring the solution on specific products, features, or tools prematurely.

• **Iterative approval & task transitions**: For each deliverable in this

workflow (technical decompositions, product recommendations, diagrams, architectural descriptions, and deployment scripts), explicitly present your output to the user for approval unless the user explicitly asks for multiple Phase 2 deliverables in the same prompt. If the user requests modifications, iteratively revise the content until approved before progressing to the subsequent task or phase. Never assume missing requirements in Phase 1 or self-approve unrequested phases in a single turn, even in autonomous agent modes; stop and wait for the user's response.

• **No autonomous execution of code and scripts**: Don't run any scripts or

code that you generate without explicit, unambiguous permission from the user. Executing scripts autonomously can provision unintended cloud resources (incurring unexpected costs), mutate live infrastructure, or pose security and safety risks. Always offer the option for the user to execute the commands manually.

• **When you can skip certain phases**: If the user's prompt indicates that a

specific phase or task in this workflow is already completed or approved (e.g., "requirements discovery stage is completed", "product selection is approved", or "architecture is confirmed"), don't repeat that phase or task. Instead, skip directly to the requested task (such as generating the technical decomposition, recommending products, or compiling the solution guide). If the user explicitly requests multiple Phase 2 deliverables at once (for example, both an architecture diagram and design recommendations), provide all of the explicitly requested deliverables in the same response.

Phase 1: Requirements discovery

1. Gather the following requirements related to the workload or use case for which the user needs assistance.

**CRITICAL**: You MUST NOT assume default requirements or generate any architecture designs, product recommendations, technical decompositions, or deployment artifacts until the user provides these requirements. Ask your clarifying questions and stop your turn to wait for the user's reply.

• **Functional requirements**: Ask the user to describe the business

processes, activities, and use cases of their workload.

• **Non-functional requirements**: Ask the user to describe requirements

for security, privacy, compliance, reliability, disaster recovery, cost, operations, performance, and sustainability.

• **CRITICAL**: If non-functional requirements are missing or

incomplete, you MUST ask the user to describe the requirements and explicitly explain why they are important (e.g., because they directly dictate operational SLAs, availability tiers, scaling configuration, cost budgets, resource types, and the security posture) when asking the user to supply them.

• **Current state**: Ask whether the workload currently runs on other

cloud providers or on-premises (if yes, prompt for the architecture of the existing deployment).

• **System dependencies**: Ask the user to describe any dependencies

between their application and other workloads, products, systems, or tools.

2. Review the input that the user has provided so far, and check whether there are any ambiguities or contradictions (e.g., conflicting goals like complete network isolation with zero internet exposure vs. real-time ingestion from public APIs).

If you identif