gcp-service-adoption
Use for Google, Google Cloud, or GCP capability adoption work. Evaluate whether a requested feature should use a Google Cloud managed service, determine whether an existing Google Cloud or GCP project
SKILL.mdregistry/skills/gcp-service-adoption/SKILL.md
GitHubGCP Service Adoption
Use this skill when a user wants to add a Google or Google Cloud capability to a project and needs more than API syntax. This skill is for service selection, project placement, governance, and rollout readiness.
Use this skill for
- Map a product need to a Google Cloud managed service
- Handle Google, Google Cloud, or GCP project/service adoption questions
- Check whether existing GCP projects can host the workload
- Decide whether to reuse a project or create a new one
- Identify required API enablement, IAM, secret, networking, and observability setup
- Define cost estimation, budgets, threshold alerts, and follow-up tracking
Do not use this skill for
- Generic GCP tutorials with no concrete project decision
- Deep implementation inside an already chosen service SDK
- One-off CLI help where only a single
gcloudcommand is needed
Workflow
1. Map the requested capability to candidate GCP services
Read references/service-mapping.md. Produce one primary service recommendation and at most two alternatives. When the primary service is one of the common profiles below, load the matching playbook as well:
- references/cloud-run.md
- references/vertex-ai.md
- references/cloud-storage.md
- references/secret-manager.md
2. Inventory existing GCP context
Read references/project-selection.md.
If the request spans multiple local repositories that each need different Google capabilities, also read references/multi-project-google-bootstrap.md.
If local CLI access exists, run scripts/gcp_inventory.py first.
If a target project and service are both known, run scripts/gcp_project_readiness.py --project <id> --service <profile> for a deterministic gap report.
If the goal is to go from capability to an initial project decision in one pass, run scripts/gcp_adoption_assess.py --capability "<need>".
Minimum inventory:
- active account
- active project
- available projects
- available billing accounts
- existing enabled APIs for the candidate target project if known
3. Decide reuse versus new project
Use the reuse/new-project decision table in references/project-selection.md. Bias toward reuse only when ownership, billing, IAM boundary, and operational fit are already clear.
4. Produce the enablement checklist
Use references/project-selection.md and tailor the checklist to the chosen service. Always cover:
- APIs to enable
- IAM roles and service accounts
- secrets and credentials
- networking constraints
- logging, monitoring, and auditability
- labels or tags for cost attribution
5. Produce the billing and budget plan
This step is mandatory.
Read references/billing-governance.md.
Read references/pricing-and-budget-apis.md when live pricing lookup, budget inspection, or budget payload generation is needed.
If a rough monthly range is known, use scripts/budget_plan_template.py to generate the initial governance artifact.
If the adoption report needs to be formalized, use scripts/adoption_report_template.py and then fill in the checklist sections from the readiness output.
If live budget work is in scope, prefer scripts/gcp_budget_api.py in dry-run mode first.
If live pricing lookup is in scope, use scripts/gcp_pricing_api.py.
6. Define rollout tracking
Read references/operations-tracking.md. The result should identify owners, review cadence, metrics, and budget alert handling.
Output contract
For each adoption decision, output these sections in order:
- Requested capability
- Recommended GCP service
- Existing-project fit
- Reuse or new-project decision
- Required setup checklist
- Cost and budget plan
- Ongoing tracking plan
Use references/adoption-output-template.md when you want a stable response skeleton.
Reference loading guide
- Load references/service-mapping.md first for service choice.
- Load the matching service playbook when the chosen primary service is covered.
- Load references/project-selection.md when project inventory or placement is in scope.
- Load references/multi-project-google-bootstrap.md when several local repos need different Google projects or credential patterns.
- Always load references/billing-governance.md before finalizing the recommendation.
- Load references/pricing-and-budget-apis.md when live pricing or live budget administration enters scope.
- Load references/operations-tracking.md when defining post-rollout ownership and review.
- Load references/adoption-output-template.md when drafting the final structured recommendation.
- Prefer
scripts/gcp_adoption_assess.pyas the default entrypoint when capability mapping, existing-project evaluation, and reuse/new-project decision are all in scope.